1

INTRODUCTION: WHAT IS A BANK?

Banks do not really exist. They are constructed. Banks are always and every-where determined, defined, and constituted by the legal permissions they are given by whichever the public authority may be: Royal Charter, Act of Congress or Parliament, or regulatory rules of thumb, preference or requirements. Differently to a physical good or type of service which does exist outside of its regulation (for example a haircut or legal advice), banking is not an activity or object on which regulation or rules can be imposed but rather one which is created by these regulations, rules and requirements. Change the rules, change the activity, change the bank.

These rule-changes – and therefore the changes to what banks do and so what banks are – happen all the time. And for various reasons. Banks, like any profit-maximizing institution, will always try to innovate beyond their explicit permissions so that they can increase their returns. By definition these new activities are outside the original scope of the banking regulation. Public authorities have a choice: to try to fill this regulatory gap or to leave it. So regulators often change rules to bring these new activities that banks undertake into their scope.

Sometimes banking regulation changes because political ideas change and so what society at large wants from its banking sector changes. Sometimes there is a greater focus on the safety of the sector, sometimes on its competitiveness, sometimes on its contribution to national production, and sometimes on the ease of access for consumers. There could be any number of political aims for the banking sector. When that changes, so do the rules imposed on banks and thus what banks can do: and what banks are.

But one thing that banking regulation does not seem able to provide is financial stability; even though this is frequently (perhaps always) stated as one of its core objectives. Despite frequent regulatory change, financial instability persists. This book argues that this is as a result of the regulatory nature of banking, of the fact that banking is constituted through regulation but operates in the market. Banking regulation is created for political reasons (in its broadest sense) but banks operate according to market principles – in other words the attempt to maximize profits within a structured market. The gap between political inputs and market outputs is where financial instability springs up.

This book argues that only political change to the ideas and objectives of banking regulation can fill this regulatory gap, not any amount of technical improvement in the content of banking regulation. As this book will argue, to achieve financial stability the objective of regulation needs to be to re-embed banking into society, and not just to smooth the functioning of banking markets or to minimize spillovers to the wider economy.

Changing what bank regulation is for, changes what a bank is: and offers the possibility of financial stability.

Why does bank regulation matter?

The real-world motivation for this book is the 2008 financial crisis and resulting regulatory treatment of banks. This is not an abstract discussion but cuts to the core of the health of the world’s financial system, economy, politics and society. Finance is unstable, regulation fails to tame it: why?

Today again, there is disruption in financial markets; regulators are unsure how to respond. This is nothing new: fintech and cryptoassets today, credit default swaps in the 1990s, and secondary banking in the 1970s all disrupted traditional banking and required regulators to think through their principles for the new market environment. The Covid-19 crisis is doing something similar for the regulatory distinction between banking and non-bank finance. Financial innovation has always caused regulatory dysfunction. The way regulators adapt to this dysfunction determines how financial innovation impacts wider society and the economy, ultimately its risks and rewards.

The financial crisis showed the world what happens when banking goes bad. The real economy is starved of funds, unemployment rises and public finances come under strain. This is because of the nature of the banking system and its interrelation with the real economy. Banks are peculiar. They are institutions permitted by states to undertake certain financial activities. Primarily they hold deposits and make loans. This allows capital to flow around an economy to be put to productive use. This is the essence of capitalism.

In return for this permission banks submit themselves to certain rules and fund state activities in certain ways. The rules, activities, and funding change over time and place, but a necessary constant is the stability of banking. As Hyman Minsky (2011) showed, finance is inherently volatile but the point of banking specifically has always been its stability. Encasing financial activity in a specific institutional form was designed to make it more dependable. It is this cloak of dependability which was rent by the crisis.

In fact, banking has become less stable over the past four decades (Barth, Caprio & Levine 2004), culminating in the financial crisis of 2008. Yet the number of bank laws and their global spread has also increased over the past 40 years (Barth, Caprio & Levine 2013). Clearly this has not led to improved bank stability. Why not? And why does bank regulation change so frequently and still not lead to stability? Under an evolutionary framework we might assume that bank regulation will keep changing until policy-makers stumble upon a stable banking framework through some combination of design, luck and trial-by-error (Veblen 1898).

There might be a systemic reason for why regulation is not able to stabilize banking. This would explain why significantly varied regulations all similarly fail to create a stable banking regime. There might, therefore, be something in the relationship between regulation and banking itself which precludes stability. In short, looking at the banking-regulation relationship as a system might yield insights. Instability is endogenous to the financial cycle (Minsky 2011), but bank regulation is about more than the market (Calomiris & Haber 2014). We need to put these two ideas together.

This book looks at how bank regulation is made and why banking is unstable. Fundamentally it argues that the banking system follows a broken form of Polanyi’s double movement caused by the difference between private (market) and public (political) institutions.

First, we turn to some of the core conceptual relationships that will be used throughout this book.

Core conceptual relationships in this book: banking and finance, regulation and stability, systems and institutions

The discussion above revolves around a number of conceptual relationships (Ball, Farr & Hanson 1989). Fundamental is the relationship between the state and the market or, in different terms, the public and private sectors of the economy. This is our core concern, as applied to the financial sector and its regulation. This relationship presupposes a number of others, however, between different concepts. These are the relationships between finance and banking, regulation and stability, and systems and institutions. The terms in these relationship are used simply but contain complexities. Most of them indeed have entire literatures devoted to them. At the outset it is worth explaining what is meant by them here. Through being clear in what we mean by these terms, and laying out how they relate together, we can lay the groundwork for the discussion to come.

Banking and finance

The basic distinction above is between finance and banking. We have already said that banking is a form of finance. This is true. But it is more than that. Historically banking is the foundation of finance (Ferguson 2008). Finance comes from banking. Banking may be increasingly marginalized within financial activity, but this is a very recent phenomenon (the past 30 years at most) and the extra, non-banking financial activity still depends on the banking sector (Noeth & Sengupta 2011).

The main problem with the word banking is that it covers a large number of different activities (Calomiris & Haber 2014). From the monetary operations of central banks, to the stock market speculation of investment banks, to the targeted investments of state-run banks, to the deal-making of merchant banks, to the savings-holdings of local banks, to the business-lending of industrial banks, the word is characterized by its versatility. Most banks do some combination of the activities schematically separated above. There are three commonalities to them all, however.

Firstly, all activities involve moving money from one part of the economy to another (lenders to borrowers, turning savers to investors, from the public to the private sector, from surplus to deficit capital holders). This is finance.

Secondly, there is a designated institutional form that structures this financial activity. This makes it repeatable, consistent and reliable. The specific form varies between the different types of banking. What is common is the fact that it is institutionalized into a bank of whatever sort. There is a non-market institutional mechanism, the firm, to manage how finance operates through the bank. This institutionalization of finance within a firm is banking.

Thirdly, the name bank carries prestige, or put another way there is value attached to this name (Calomiris & Haber 2014). Over time this has often become a reputational value, but it is also more specific than this. To become a bank you have to be licensed as a bank. It is a protected term. This means it is a privileged term. This gives banks an advantage in the market place because they are permitted, by the authorities over the market (otherwise known as the state), to conduct certain activities in certain ways which non-banks are not allowed to do. It gives banks a certain monopoly power over areas of financial activity. Banking could be conducted in many ways – indeed, as the idea of shadow banking shows, it is – but only through licensing does an institution become a bank.

Thus according to these three commonalities, we define a bank for our purposes as a privileged institution licensed to perform specific financial functions. Practically speaking, we will limit this definition to commercial banks – those that take deposits from and lend to business and the public – since they have the most effect on the wider economy through their systemic interactions with the real economy and citizens. Because of interrelation with the economy and society politicians are most worried about this kind of banking and legislate it the most and most specifically. These banking functions are also the most long-lasting and prevalent.

Banking is not, however, the only type of finance. Today we might think of non-bank financial institutions (NBFIs) or new fintech firms. Stock markets are perhaps the most obvious form of financial activity not involving banks. Indeed for most of history retail banks were forbidden to operate on stock markets (Kynaston 1994). These represent a market-based way of performing bank-like functions: moving surplus capital into projects which require it, in hopes of a return. Bond markets, corporate and sovereign, also form investment opportunities as do commodities and other exchanges. What these all have in common is the market mechanism for moving capital through the economy according to a regulated structure – the exchange – rather than through in-house decisions. They are market- not firm-based. More recently, instruments such as money market funds (MMFs) have also formed to create a market-mechanism which can mimic the savings-function of a bank. There are many other ways of creating financial returns – making a profit by managing, moving, or lending money – without being licensed as a bank. Indeed as financial economies develop it is precisely in this area that innovation occurs, with banks themselves partaking more and more of this non-bank finance; as far as they are allowed to. That permission is our very area of focus.

Banking was the original form finance took in the modern era, its management of the money in an economy led it to develop financial markets, while the institutional form itself – and the permissions, protections and limitations that go with it – incentivized cheaper, easier, less-overseen forms of finance and a corresponding loosening of banking regulation (Ferguson 2008; Boyer & Kempf 2016). Banks are the core of the financial system and its main intermediary with the real economy, but they are not the only financial actors. But the primary focus of financial regulation still comes in the form of banking regulation.

Regulation and stability

This requires us to have a clear idea of what we mean by regulation, a term no simpler than banking. The crucial assumption in the initial discussion above is the one between bank regulation and stability. There are many forms and aims of regulation (Llewellyn 1999). Put simply, and very broadly, we use the term regulation to mean the action of the state in and on the market. As the word itself shows, it is an act of control. Ideally it is an act of control to fine-tune or optimize otherwise wild activities. We have deliberately chosen a very wide definition of regulation, as a concept, because the methods of regulation can be, and are, so varied. The very aim of this book is to understand how the state interacts with financial markets. Necessarily there are many possible ways it could, and indeed does, do this. Therefore we need a concept and term broad enough to encompass all these different interactions and yet still consider them together at the same time. The word regulation does that.

The problem with the word regulation is that it means something specific too: the specific tool or instrument of the concept regulation as discussed above. A regulation is the method of regulation, or at least it is one of them. In this sense a regulation would be a law, ruling, or requirement as determined with varying legal force by a state body, to which private market actors have to adhere. A regulation is therefore just one form of regulation as we understand it. In practice this ambiguity is not so severe, since it is nearly always clear from context and discussion what is being meant.

One further complication is in terms of financial market regulation, where regulation tends to mean prudential regulation and is linked with the supervision of banks (Dewatripont & Tirole 1994). Prudential regulation is a specific set of capital rules for banks, as determined by banking authorities (prudential regulators and supervisors), to ensure their stability, ability to operate under adverse market circumstances, and to protect unsophisticated consumers and the tax-payer in general from bank failure. In modern times the main prudential rules come under the Basel framework and are now generally coordinated internationally, although there still continue to exist significant country differences. Over the past few decades prudential regulation has become the principal form of bank regulation. But it is important to be clear that when we discuss bank regulation it means the overall attempt by the state to control how banking works, not just prudential regulation. To repeat, often and increasingly bank regulation means prudential regulation, but they are not the same thing and are kept conceptually separate.

This distinction matters when it comes to the idea of stability. Prudential regulation (and supervision) is concerned specifically with bank stability at the institutional level. It forms a set of specific, legalistic, concrete rules which banks must abide by, with the intention that this will limit the likelihood of a bank failure, costs to its stakeholders, and impact on the real economy and government finances (Dewatripont & Tirole 1994). In terms of bank regulation, stability means something else or rather something more. It must mean stability of banking as a whole, since the stability of the sector is a necessary component of whatever aims and interests the state may have in banking (and financial markets more generally). This is because, in banking, stability cannot just mean the stability of individual banks: they must be allowed to go bust, indeed most efficient financial theory would suggest that poorly managed banks must go bust (Malkiel 2005). Equally stability in banking cannot just mean that banking activity will always grow in an economy with no downturns or periods of financial repression or scarce, expensive credit. Again, sometimes this is to be expected and welcomed in an economy, however many people lose out. It is perfectly acceptable for there to be a business cycle in banking.

What stability has to mean in the context of banking is systemic stability, the stability of the banking system, not just of individual banks or of their activities within the system. This is a form of catastrophic or crisis stability, the ability for the banking system to weather economic and financial downturns and the failure of individual banks. We take bank regulation in its broad sense because it is only in the broad sense that states develop banking systems, allow banking institutions to operate and then attempt to control these banking activities through more specific (often prudential) regulation. Bank regulation refers to the entire attempt by states to have a stable financial system in their economy. Again, the reason stability is a fundamental aim of bank regulation is because whatever other aims states may have for banking – tax revenues, cheap credit, economic efficiency, financial redistribution (in either direction), international integration – all presuppose a stable banking system, in the sense of not prone to catastrophic crisis, regardless of what happens to financial activity as such or individual banks. A stable system is the base on which all banking is built, and regulation the means of securing it.

Systems and institutions

Discussing the banking system necessarily requires us to have some idea of how systems operate. Simply put, a system is an environment in which elements perform interactions for a purpose (Meadows & Wright 2008). There are thus three equal components that make a system: elements, interactions and purpose. The key point is that the system causes its own behaviour; that is, it is latent in its structure. A system can best be seen as a persistent pattern. The relevance of this is that change in one of the three parts of the system affects all the other parts (Jervis 1999). Our definition of regulation means the state can, and does, influence all three aspects of the system and thus the size, functioning and success (on its own terms) of the system as a whole. However, as the state changes, or gets involved with, one part of the system, it wittingly or not changes the others.

To make this discussion a little less abstract, we should turn to the banking system itself. It is made up of banks (elements), their activities (interactions), and the system’s purpose which is to move money from where it is not needed to where it is for a financial return. This should improve the overall productivity of the economy (King & Levine 1993). At least to start with this takes place all within a domestic economy context. What we see over time is the state changes both the economy’s openness to the outside world (the actors), and also what activities banks are allowed to undertake (interactions). In so doing, we can assume the state changes the system’s purpose: the balance of economic efficiency of banking to individual actor’s interest-gaining, as well as the geographic scope of the system. All these changes will impact the system’s stability, and not necessarily in ways that were intended or foreseen. One of the characteristics of systems is that specific changes to one aspect of the system can lead to wildly exaggerated and perverse outcomes in another (Meadows & Wright 2008). Assuming that a banking system starts off in an equilibrium state – that is, where the three parts of the system cohabit positively, without forcing changes in each other – then we might suppose that state involvement in one aspect might cause a shock which knocks the system out of an equilibrium. There is no particular reason why we should assume the system will then find another equilibrium, let alone a satisfactory one (Baumgartner & Jones 2009). Looking at what has happened historically is the only way to develop a generalizable insight.

In the real world, these diverse components of a financial system are all different types of institution. Institutions are long-lasting, analysable since non-arbitrary, and repeating. Individual actors are important, especially when looking at the historical record, but the way to generalize answers out of individual action is to see it in terms of the institutions that action represents. The aim of an institution is to create a certain reliability in individual action by making humans conform to an established pattern (North 1994). Put another way, an institution is a humanly-devised constraint which patterns human action (Peters 1999). The repetition gives the point. Human action is both the input and the output of an institution, which thus, in anything other than a static analysis of one point in time, becomes to some degree a feedback loop. Humans construct institutions, which then change human action, which then alters the context in which institutions operate or causes humans to alter the institution. According to this understanding, institutions also exhibit path-dependence which means that the initial form of the institution will persist until it is knocked off course and changed by events or deliberate action (Goodin 1996). Institutions exhibit inertia in the physics sense. They therefore not only structure human action but also make it persist.

In Chapter 2 we will look in depth at the literature surrounding institutions, and how an institutional analysis can help unravel some of the issues this book deals with. At this point, though, it is worth flagging up the very different logics behind private and public sector institutions (Coase 1937; Posner 1974). At core, private institutions aim to profit-maximize while public ones aim to produce publicly-desired outcomes. This second point is not simple: it depends on where the institution derives its legitimacy from, who controls it, and who will benefit from it. Looked at another way, public institutions involve the political process of hierarchy, popular support and control, while private institutions operate through firms operating in markets, according to available information and price signalling. The interaction between the two types of institutions is one way of understanding the banking-regulatory system.

As discussed above, regulation is the action of the state in the market. In short, it is the interface between public and private institutions in the banking system. The institutional actors of the banking system are not just the private institutions of banks but also the public institutions of regulation (government, regulatory authorities, central banks, etc.). By examining banking regulation directly, we will therefore have to understand how these two different kinds of institutions interact in the banking system, and then see what the systemic effects of this interaction are.

The political economy of states and markets

So far this discussion is very firmly on the ground of political economy literature. It looks at how institutions interact within systems according to political and economic processes. For such a broad question it is no surprise that a number of ways have been devised to answer it.

The aim now and in the next chapter is to draw these general concerns, questions and approaches more specifically into the realm of banking regulation. Later this will be done by moving the discussion to a specific time and place – the UK between 1970 and 2020 – which will make the conversation more concrete. Before that we will look specifically at the literature on banking and financial regulation and on financial market activity itself.

Banking and finance, and their regulation, are special sectors of modern capitalist economies because of their size, dominance and centrality. The financial crisis of 2008 did not just knock out a wealthy sector of the developed economies, but severely damaged all sectors of these economies and government finances. This is to be preferred to a general socio-economic (and indeed political) collapse as happened in the 1930s, but it has left a toxic legacy: both in financial markets and further afield. The financial aspects of Covid-19 and the UK’s withdrawal from the EU are also likely to have significant impact on the economy and state, at least in the UK and EU. Examining banking and its regulation is worthwhile for itself.

Banking and regulation

The institutional form of a bank (what a specific type of entity is allowed to do within a financial market) represents the political compromise of a given time and place, and one that can therefore change frequently. It is not stable and only indirectly reflects market developments. The bank itself originates from permissions granted by the state, for the state’s own reasons. Banks do still of course operate in markets as economic actors – they attempt to maximize their profits. An implication of the imperfect alignment between the politically-derived form of banking and the business aims of banking is the inability of bank regulation to tame financial instability, its purported aim. Regulation is understood in the broadest sense of the state’s attempt to control how actors undertake financial activities, not solely in the modern prudential sense – which is only one version of regulation.

Learning from Polanyi: the relationship of state to market

First it will be helpful to explain how states constitute markets more generally. This is Karl Polanyi’s key insight (Polanyi 2001). Many scholars have built on his work in the subsequent decades and used it to explain a great number of things. Later in Chapter 2 we will turn to them. But here at the start it is useful to go back to Polanyi’s original framework and see how free markets are in fact constructed by states.

Polanyi argues that the idea of a self-regulating market is a myth created for political reasons. Through a discussion of English social and economic history, he shows how the market is one part of wider social relationships; and one which requires ongoing state intervention to function appropriately. As Stiglitz puts it in his introduction to the 2001 edition of Polanyi’s The Great Transformation: it requires statecraft and repression to impose the logic of the market and its attendant risks on ordinary people. Furthermore, over time the fictitious commodities of labour, land and capital gained widespread acceptance as seemingly natural areas of market transaction: that is, traditionally