Location: Westminster Business Forum on Banking reform in the UK, Glaziers’ Hall, London

I’d like to thank the Westminster Business Forum very much for giving me the opportunity to address this distinguished gathering. Just to be sure, I am not stating any formal positions of the FCA in my talk today.

As most of you know, though, the FCA is taking initiatives on culture in UK banking, so I will briefly summarise the state of play on these.

I will then speak from an economic perspective on why good conduct culture in banks is important for financial stability and growth. I will argue that our regime helps this culture to develop and that societal pay-offs from this culture are increased by our actions to promote competition.

Finally, and perhaps most importantly, I will discuss what economics and related disciplines tell us about establishing and maintaining an appropriate culture in organisations.

To explain the last point a little. There is a long history of apparently clear principles of financial conduct regulation such as ‘treating customers fairly’ and of firms signing up for these principles with excellent intentions. Yet there is also a long history in UK financial services of widespread breakdowns in compliance with such principles. I infer that compliance in practice is harder than one might suppose it to be. So I will seek to offer insights from economics and related disciplines which may help firms with this practical problem.

To do this, I will draw on two Occasional Papers which will shortly appear on the Insight pages of our website. These pages are the FCA’s ‘home for independent opinion’. My remarks today are part of this independent opinion.

The FCA and UK banking culture

So first I address the state of play with respect to the FCA’s initiatives on culture in UK banking. Culture is a priority for the FCA, one of our seven business plan priorities for 2016/17.

We believe that poor culture played a significant part in the financial crisis and that it is a root cause of many failings at firms. Thus culture is both a major driver and potential mitigant of risk. Our ambition is that firms’ senior management lead and foster a culture that has the fair treatment of customers and market integrity at its core.

Hence the main aim of our major initiative, the Senior Managers and Certification Regime or ‘SMCR’, is to drive cultural change. It does this by making senior managers accountable and by applying baseline standards to all financial services staff.

For the banking sector, we recently published policy documents, including rules on regulatory references, various consultation papers and a discussion paper on the application of the SMCR to the legal function. We also published feedback on firms’ implementation of the SMCR. We found that:

• The great majority of firms outside the credit union sector had identified Senior Managers and allocated Senior Management Functions and responsibilities.
• There were some common issues such as insufficiently senior individuals identified for SMFs, not identifying all areas of responsibility, and highly variable maps of responsibilities.

I note for any non-banks present that:

• The Bank of England and Financial Services Act 2016 extended the SMCR to all FSMA authorised firms. 
• Treasury is looking for implementation from 2018. 
• We are working on proposals that will meet the needs of firms, drawing on insights from relevant supervisors and meeting trade associations for discussions.
• Important challenges are the diversity of types and sizes of firms, and our determination to achieve a proportionate approach.

Banking culture and financial stability

I said that my second topic would be to speak from an independent, economic perspective on why culture in banks is important for financial stability and growth and the valuable role of regulation in this. I do not want to labour the obvious point that banks with a dubious culture might focus on creating short-term profits to pay bonuses or to look successful while initially not recognising hidden risks and losses and ignoring harmful impacts on counterparties and society more broadly.

Nor do I want to labour the equally obvious point that this behaviour can ultimately lead to the collapse of banks, and to a sometimes very difficult and always very undesirable public policy choice: is it cheaper to let a bank fail or to bail it out?

In fact, while the salience of the enormous numbers associated with financial crises has caused disquiet over the years, I am going to argue against views such as those of John Adams, the US president from 1797-1801. He claimed that 'banks have done more harm to the morality, tranquillity, and even wealth of this nation than they have done or ever will do good.”

Here then is a more interesting and distinctly different, positive story. This is a story, in my view a true story, in which the social benefits of much of the financial sector are enormous. The problem is that these benefits are difficult to see because one cannot easily envisage the so-called counterfactual: what our society would be like without modern finance. On this, perhaps one should reflect on the World Bank’s insistence on conditions for capital formation as a critical driver of transition from developing economy status. Or just consider the huge change in living standards in countries developing their financial systems for the first time.

In brief, the positive story is as follows. Finance assists growth. Growth assists prosperity. But only ‘good’ finance assists stable growth.  And finance is most likely ‘good’ when culture is ‘good’.

Much of the underlying empirical and theoretical literature that supports these propositions is by Ross Levine and his collaborators. They find that finance greatly affects the economy as a whole through the choices it makes in allocating capital. Specifically, good choices create prosperity. Let me explain more.

Those who have funds they wish to save for the future want to do so with as high a return as possible given the level of risk (and time commitment) they are willing to accept. Those whose business ideas require capital need to access capital markets and attract funding in the form they require. The capital markets need to allocate capital to those whose ideas are better than the alternatives.

When markets work well, savers’ money is matched with the best investment opportunities. Effective companies get funds that allow them to invest to start, transform or expand their businesses. Products that insure their business risks are available. Together, this means that more efficient and desirable products and services can be produced and sustainable jobs are created.

It also means that consumers are able to borrow to an appropriate extent in order to buy housing and consume products they need. Further, the return from these activities rewards savers for providing capital and allows them to spend more in the future.

Through this efficient allocation of capital and risk, the economy grows in a stable and long-term way. It is worth adding that for financial services firms to deliver this positive effect on the long-run rate of economic growth, they need to avoid imposing prohibitive transaction costs on the processes involved. Moreover, financial innovation is needed to support technological change and its related economic growth.

Culture in financial services firms is critical to their ability to deliver the benefits I have just described. A firm with a good culture is likely to allocate capital well; that is, on the merits of the business cases. A firm with a good culture is likely to monitor the use of the capital it has allocated, so as to check that its clients get the stewardship and, subject to shocks, the outcomes they expect. A firm with a good culture is likely to innovate to the advantage of its clients and to lower transaction costs. All of which tends to lead to growth.

In what I have just said, the terms ‘culture’ and ‘good culture’ have not been carefully defined. For some of you I suspect that the brief description of regulatory initiatives with which I started this talk will suffice as definition. Or you might be well-informed through reading some of the fascinating articles on culture that appear in the McKinsey Quarterly or in the proceedings of the Caux Round Table. But, to make sure that we are on the same page, I will say a little more about this.

What does ‘good culture’ mean?

By ‘culture’ I mean a combination of shared or, at the very least, dominant beliefs, values, conventions and habitual behaviours within a group. For example, it might encapsulate how the members of a group relate to each other and view outsiders.

An important point here is how one thinks of the term ‘group’. It does not have to be the firm as a whole. In fact, there is academic work on how challenging ‘sub-cultures’ can be in finance. As somebody who once sat in an uncomfortable role between Samuel Montagu and its clearing bank owner, I can personally vouch for these studies.

‘Good’ culture within a firm could involve a shared belief – and pride - in the importance of delivering value for clients, emphasis on long-term relationships with clients and business partners, and a habitual practice of acting with honesty, prudence and personal accountability within the company and with outside stakeholders. From this perspective, it is easy to see, I think, how ‘good’ culture helps a firm make better decisions in allocating capital, thereby supporting the economy as a whole.

As already intimated, none of this means that finance always has these great qualities. Some economic literature shows how bad banking culture can lead to self-serving innovations that dupe the unwary, create instability and waste societal resources. The question is whether statistical analyses of finance and growth suggest that such behaviour is the preponderant case. Arguably, they do not.

Similarly, on growth and stability, there are differing views but quite a strong consensus on their being generally positively correlated. The IMF has published papers on this. The European Commission has issued a Stability and Growth Pact.

I am going to make one last point on my second topic and for me it is particularly significant. This point concerns the importance of conduct regulation to growth and stability.

As is well known, the FCA is represented on the UK’s Financial Policy Committee. What is perhaps less appreciated is quite how important this is.

The argument partly follows from what I have already said. If ‘good’ culture is an important determinant of whether we have ‘good’ finance that leads to growth and stability, and if the FCA drives ‘good’ culture in activities relevant to capital allocation, which I strongly believe it does, then the FCA’s activities contribute to growth and stability.

And there is significantly more to say. It is possible to conceive of financial markets in which firms have ‘good’ culture and yet the markets still do not work well. We might think of ‘good’ culture as a necessary but not a sufficient condition for efficient allocation of capital.  In this case, capital could still be misallocated. Growth and stability would then be less than they could be.

For example, costs of financial intermediation might be excessively high as a result of a concentrated market structure. Or traditional practices or rules of various kinds might make markets inefficient economically.

It is therefore of critical importance that the FCA’s overall objective is to make markets work well. It is also important that the FCA has an operating objective to promote competition, and a competition duty which requires it, other things equal, to reach first for competition tools when it needs to act. Together, these obligations can drive a programme of regulation that will enable markets to allocate capital better and generally work efficiently, providing appropriate access to demand and supply sides.

Detailed thinking about this can be seen on the Insight pages I mentioned earlier, in the form of our publication ‘Economics for Effective Regulation’.

Economics and culture

Turning now to my third and final topic: what economics and related disciplines tell us about establishing and maintaining an appropriate culture in organisations. Again, what I am going to say is independent opinion not a formal position.

The difference between the level of tolerance of non-compliance in financial services regulation and in some other regulated fields is very striking. In aviation, trains, nuclear, oil exploration and so on, there is near zero tolerance of non-compliance. On the other hand, in financial services we have sometimes seen systematic non-compliance.

I became more intrigued by this difference when some non-executive directors of financial firms told me the link between a staff member being disciplined by the FCA and breach of firms’ internal disciplinary codes. Specifically, in some cases there was absolutely no link.

Economically, this seemed surprising. FCA disciplinary action should be bad for reputations and revenues, and fines harm profits further. The management should, in principle, want to keep reputations and revenues high. Why then did they not always discipline the staff responsible?

The chain from rules to compliance should be simple. That is, economists might think of compliance as involving two principal-agent arrangements.

Now I realise that I have just slipped into economic jargon. So let me quickly offer a translation into management speak. Businesses care about culture because, along with so many other things, including information and psychology, it affects the decisions that people make. As Doshi and McGgregor put it in ‘Primed to Perform: How to Build the Highest Performing Cultures Through the Science of Total Motivation’ (or ‘ToMo’):

‘If you had an organization where all people did was robotically execute some plan, then culture wouldn’t matter.’ 

Back to economics. In the first principal-agent arrangement, the directors of the firm act as the regulator’s agent in overseeing business in accordance with the rule book. In the second arrangement, the staff act as the firm’s agent in executing the business in accordance with the directors’ and by extension the regulator’s requirements. What could possibly go wrong?

A lot, of course, as even basic knowledge of economics would suggest. Without additional constraints, there is little reason to believe that the ‘agents’ in these relationships have the same incentives and objectives as the ‘principals’. And since the ‘principals’ cannot monitor their agents’ actions perfectly, and certainly cannot control them directly, it is not surprising that the chain does not always deliver compliance with our rules. Sometimes the parties in the chain do want different things.

Economists have traditionally thought this problem best addressed by trying to change incentives. This means changing the private benefits and costs of particular courses of action for the agents so that they actually prefer to do ‘the right thing’. In regulation or law enforcement (and, indeed, in firms’ own internal monitoring) this is often attempted by aiming to create ‘credible deterrence’. This could mean raising penalties for misconduct and/or raising the probability of its detection so that the expected cost of the wrongdoing to the agent becomes higher than their private gains.

This level of deterrence is often difficult to achieve in practice. Moreover, if as already mentioned, decisions are materially influenced by culture and information and psychology, and by further factors such as social context, then it is unlikely that they could be sufficiently understood solely through neoclassical economics nor even through behavioural economics with its combination of information economics and biases derived from a relatively narrow area of psychology.

So we decided to explore what fields outside traditional economics and financial regulation can help promote compliance. To do this we requisitioned two pieces of analysis which will shortly appear on our Insight pages.  One is a review of existing economic literature on behaviour in organisations and what it means for compliance. The other seeks to draw lessons on compliance from the world of tax.

Let me start by talking about the second of these papers.

Why look at tax in the first place?

In some ways, tax authorities have a harder job than financial regulators in promoting compliance. First, tax payments are a pure cost with no direct benefits for businesses making the payments. Second, tax authorities know on average less about most tax payers than financial regulators know about regulated firms. Moreover, it was interesting to learn that in the UK HMRC had instigated a special code of compliance for banks, one requiring banks’ boards to sign up to some very specific undertakings. This suggested that tax compliance in banking might be especially difficult to achieve, further implying that we could draw lessons from HMRC’s deliberations on the issue.

We therefore requisitioned two leading academics on the economics of tax systems, Gareth Myles and Chris Heady, to write a paper setting out some key strategies that tax authorities use to promote compliance and their implications for financial regulation. I emphasise that this paper, like our own to which I will come later, is about improving compliance, a goal to which culture can contribute. It is not about prescribing culture.

Here are some of the things we learnt from Myles and Heady.

First, as suspected, there are many parallels between tax authorities and conduct regulators. In fact, some of the initiatives in HMRC to promote compliance are very similar to those in financial regulation. For instance, the creation of the HMRC Large Business Service for individual relationship management with the largest corporates is similar to the FCA’s ‘fixed’ relationship supervision. Similarly, the introduction of Senior Accounting Officer rules in 2009, which assign personal responsibility and non-compliance penalties to a senior manager in a taxpaying firm, has many parallels with the FCA’s Senior Managers Regime for financial firms. There are clearly lessons to be learned both ways.

Second, tax authorities undertake genuinely random audits of firms’ returns, in addition to targeting specific firms identified through sophisticated statistical methodologies. Using a range of data about firms that were and were not found to be engaging in tax evasion, these methodologies identify combinations of characteristics that signal whether a firm whose tax behaviour is unknown is more likely than most to be a tax evader. These predictive analytics are considered materially to increase the pay-off from the scarce resources available for auditing activities. Similarly, the US securities regulator, the SEC, has already used a system like this (text analytics) to target firms that misreport their financial condition. Overall, the results were so positive that the SEC is now deploying such techniques in a wide range of applications.

Thirdly, the authors highlight an important difference between tax administration and financial regulation. Broadly speaking, only the aggregate tax payment of the firm matters. On the other hand, in financial services mis-selling is likely to occur at the level of individual client transactions. For the kind of detection described above regulators likely need to gather and use detailed, transaction-based data.

Fourthly, the introduction of a ‘presumptive’ ban on tax avoidance schemes is interesting. This deals with attempts to bypass the rules rather than violating them explicitly. Under the General Anti-Abuse Rule introduced in 2013, a scheme which reduces tax due is deemed to be abusive tax avoidance unless there is some commercial justification for it other than reducing tax liability. This approach might be helpful in financial regulation in deterring abuse of unintended loopholes. An example could be when firms react to a ban or restriction on a particular class of products by introducing new products that legally do not quite fall under the definitions in the relevant rule but have the same economic purpose as the banned product.

Fifthly, HMRC has collaborated with professional bodies such as those which oversee accountants and lawyers to set mutually reinforcing reporting requirements with respect to tax schemes. External advisers on such schemes are given an incentive to report them to HMRC through the similar reporting requirement imposed on professional staff in banks who work on such schemes. And vice versa.

Finally, while Myles and Heady focus on improving compliance through improved deterrence, they observe that even tax evasion can bring psychological costs of non-compliance whether or not it is detected. Feelings of shame and guilt are a psychological ‘punishment’ for tax evasion. And they are likely to be stronger in societies where benefits of tax compliance are widely appreciated and evasion is considered unacceptable. When strong, these feelings reduce tax evasion. These ideas are connected with the notion of ‘tax morale’, something which authorities seek to build, as it leads to strong voluntary compliance. This neatly leads us back to the role of culture.

Behaviour and compliance in organisations

The FCA’s own upcoming Occasional Paper in this area is called ‘Behaviour and Compliance in Organisations’. It focuses on what we regulators can learn about improving compliance by combining insights from behavioural economics, sociology and psychology with more traditional economic thinking on credible deterrence. It also considers what firms themselves can learn from these sources about encouraging compliance. We take it for granted, of course, that this activity by firms is in the context of business models that sustainably provide an acceptable rate of return to the providers of capital.

The material we cover is very broad, so I will only highlight a few observations from this paper here:

First, it is not just the size of the probability of detection and the penalty for non-compliance that matter. It is also their salience and vividness to the decision-makers considering engaging in misconduct. So, for instance, there might be a case for targeting more penalties at individuals directly responsible for the wrongdoing rather than at the firm. This makes the risks more ‘personal’ to other people who might be tempted by non-compliance.

Second, clear, salient and regular regulatory communications about penalty actions are likely to increase the impact of the probability of getting caught and punished. In my view, this contains a lesson for a well-intentioned firm that wants to deter misconduct among junior staff. To increase the salience of potential penalties, these firms could make sure that whenever a regulatory fine or penalty is published – even against someone else in the market – its own staff working in the relevant area are made aware of the enforcement action. The effect of this could be enhanced by strong discouraging messages from local management and/or an invitation to whistle-blow.

Thirdly, morality is a very important factor in decisions to engage in misconduct. Almost all people value being able to maintain a view of themselves as a decent person. So, it is much easier for someone to engage in harmful non-compliance when there is a lot of mental distance between their action and the detrimental consequences to the consumer.

Let’s make it clear with an extreme example. Consider pressing a button on the screen to give someone else an instruction to add a terrible product to a consumer’s basket of investments. Now consider approaching the same consumer on the street and stealing cash from their hands. Most of us somehow find the former less viscerally wrong than the latter. Yet the former will substantially erode the victim’s retirement savings, possibly impoverishing them. While the latter will likely do little more than deprive the individual of some small, immediate consumption.

The policy question raised by this is whether culture can make the former as abhorrent as the latter. For instance, firms can help employees reduce the mental distance from the victims of misconduct by reminding them – regularly – about the effects of their financial products on the lives of their customers. Or they can create accountability structures where there is as little diffusion of responsibility as possible. Then it becomes difficult to distance oneself mentally from a non-compliant decision because it is not quite anyone’s fault.

Fourthly, as I have already mentioned in my examples, the literature we reviewed identifies a large role for culture in compliance. Anthropologists tell us that people are very influenced by social norms in the group around them and often change their beliefs and possibly even preferences to conform. It is therefore important to have consistent positive values, about which staff are regularly reminded, and to reward behaviours that align with these values. In fact, the ease with which any individual employee can mentally justify misconduct can also be shaped by things as simple as the company’s beliefs and internal language around business objectives and consumers. For instance, whether the latter are colloquially referred to as ‘valued customers’ or ‘money bags’ or, in at least one famous case, ‘muppets’. Clearly this is not a matter that is entirely outside the control of senior management.

Overall, the findings of the upcoming OP suggest that while credible deterrence of non-compliance is important, many other factors can play a large role: firms’ internal communications, corporate policies, job descriptions and remuneration structures can all shape culture and compliance decisions. So can regulators’ choices in the targeting of enforcement, publicising their activity and reinforcing good practice.

Indeed for firms wanting to change their culture a valuable approach may be to recognise that culture is a product of a wide range of contributory factors and then decompose it into its main drivers so that the role of each can be considered and developed. During City Week 2016, Andrew Bailey summarised the drivers as follows:

‘The stance and effectiveness of management and governance, including that well used phrase “the tone from the top”; the structure of remuneration and the incentives it creates; the quality and effectiveness of risk management; and as important as tone from the top, the willingness of people throughout the organisation to enthusiastically adopt and adhere to that tone.’

It’s time to conclude

I have tried to give a broad conceptual understanding of factors driving culture and how one might go about influencing culture. My reasons for doing so are that broad understanding helps specific, practical decisions and increases the chances that these decisions are consistent with each other; that simple observation of culture in practice is not the same as understanding it; that all of us can learn by evaluating and aggregating the ideas to be found in high-quality papers; and that culture is a cross-disciplinary issue.

The last point is worth expanding a little. Most of us are not polymaths. Our understanding of the world tends to be rooted deeply in the science in which we majored and/or in which we work. So a topic like culture which cuts across many traditional scientific disciplines can easily be misunderstood as individually we perceive it from just one point of view. It may therefore best be addressed by multi-disciplinary project groups.

Pursuing this thought, let me finish on a practical point. The Centre for Evidence-Based Management is not as well-known as the excellent consultancy I mentioned earlier. It is a not-for-profit, multi-disciplinary network of researchers and others that aims to help managers and organisations make better decisions. Its approach certainly aligns with my theme that broad, new thinking about culture can be valuable. It has carried out at least one very interesting and original project on culture with a major UK bank, and perhaps it can help others take a fresh look at this challenging issue. Needless to say, I am not giving advice!

I hope that you found something useful in my remarks today and I am happy to take any questions.