Banking: Intermediation or Money Creation
S. Das, Roundtable Wrap-up
August 3, 2020
Sannoy Das, Harvard Law School
In this brief post, I attempt to summarize the main themes that emerged from the Just Money roundtable on banking. In ten blog posts between January and March 2020, before the coronavirus pandemic turned the world upside down, our contributors set out to answer the question of what it is that banks do; and more importantly, why understanding what they do is so important. The prompt for the roundtable asked our contributors to discuss the two known paradigms about how banks work: on the one hand, the view that banks intermediate funds between savers and borrowers (often the ‘orthodox’ view), and, on the other, that banks create money through lending (the ‘heterodox’ view). Most of our contributions also address, some more directly than others, what is at stake in choosing between these views about banking. Our contributors, as we will see, disagree on how and whether the choice of paradigm dictates normative choices about the regulation of financial market entities, and, perhaps more fundamentally, whether the debate over views about banking is enmeshed in a broader ideological struggle about the relationship between public power and economic activity.
Marc Lavoie, in the first contribution to the Roundtable, answers the prompt most directly, offering a clear statement of the heterodox ‘money-creation’ view and its venerable historical antecedents. For Lavoie, this view puts the difference between the banking system and the ‘shadow-banking’ system in sharp relief – and thus offers a path ahead for regulators in handling crises in financial markets. Howell Jackson on the other hand, argues that policy makers ought not to abandon the orthodox intermediation theory, because crises such as bank runs are, in his view, events of dis-intermediation. For Jackson, the intermediation theory persuasively describes the behavior of practical bankers, always interested as they are, in acquiring their deposits. And this, in my view, opens up an interesting question: in picking between these two theories, should we look for one that best approximates the experience of bankers or one that best captures some fundamental dynamic about the banking system? The case that Jackson and Charles Kahn make is that intermediation view describes the constraint that bankers face in lending. Others, as we will see below, believe that this constraint is a tangential consideration in picking the better theory.
Sir Paul Tucker, formerly of the Bank of England, argues that there is nothing heterodox about the view that bank deposits, which are money, are created by banks in their act of lending. But for reasons of mathematical simplicity, the idea that banks intermediate funds according to rules of fractional reserve banking and the related idea of the ‘money multiplier’ are embedded in economics textbooks. These concepts may have made better sense in a context where central banks regulated a bank’s ability to create money by controlling the supply of reserves, not in the (more current) context where the primary policy tool for central banks is regulating interest rates. Tucker’s bigger takeaway, however, is that debating theories of banking is to put the cart before the horse. What matters more is to craft policy measures that keep the monetary system stable, given our agreement that banks affect the money supply and are volatile institutions. In a stable monetary system, where bank deposits are safe assets, banks will be able to drive money supply by extending credit.
But others take the flip side of Tucker’s cart-and-horse. They argue that the view that we adopt about the nature of bank activity shapes our approaches to bank regulation. For Morgan Ricks, the distinction between the two views of bank operations influences the question of how to regulate financial entities at a normative level. In his view, the ‘intermediation’ approach primarily serves to blur the distinction between banks and other financial institutions in service of two conservative policy outcomes. First, the intermediation paradigm suggests a deregulatory bias for banking, since it appears similar to other financial entities and second, the paradigm invites other financial market participants, such as hedge funds, to stake claims for support from the central bank. Stephen Marglin echoes Ricks by noting the mediation orthodoxy is ultimately an “ideological tool in the fight for deregulation.” This is why it remains in textbooks, despite the flawed underlying assumption that banks play a passive role in the economy. For Marglin, the difference between the two views on banking does not turn on the question of how constrained bankers are in making loans (indeed, he admits that reserves can be constraining). Instead, it turns on understanding whether reserves or the money multiplier are exogenous to banking activities – a premise he rejects. Reserves and the money supply are determined by the actions of bankers, which explains the frequent situation of excess reserves in the banking system.
One broad takeaway from Marglin’s contribution is that how we pick between the two ideal-type accounts depends on some broader conceptions of the field of macroeconomics. This much is evident in the disagreement between Charles Kahn on the one hand and Kumhof and Jakab on the other. Kahn’s argument is that while it is true that banks create ‘money’, this is predicated on defining money to include bank deposits and to exclude many other forms of credit. Conceptually, for Kahn, the important consideration is not whether an asset (debt) is money, but whether it is liquid – for if it is liquid, it can function like money. If the more relevant macroeconomic variable is liquidity, then banks are similar to many institutions that issue, or are capable of issuing, liquid debt. Without the backing of deposit insurance, or implied too-big-to-fail protections, banks would be constrained in issuing debt, just like other financial institutions. Thus, Kahn argues that without reference to the macroeconomic context in which particular banks operate, it would be misleading to state that bankers can simply issue debt with the ‘stroke of a pen.’ To Kahn, given that there is nothing fundamentally exceptional about banks, the intermediation paradigm is a good fit.
Michael Kumhof and Zoltan Jakab disagree pointedly. For them, focusing on liquidity as the appropriate marker for differentiating (or obliterating the difference) between banks and other financial institutions is a mis-step, one that will undermine our understanding of crisis and regulation. Instead, the fact that a banker can create debt ex nihilo and a non-banking financial institution cannot (all conditions being equal), is critical. It explains why bank balance sheets are more fragile and why the increase or contraction of money flow is much faster in response to slight perceived changes in macroeconomic conditions. Their point, of course, is not to deny that bankers are constrained in how they make out loans, but that the constraint is ultimately based on the banker’s view of profitability, which is theoretically (yet significantly) distinct from the need to first have attracted deposits.
I noted above that Morgan Ricks approached the debate between the two views of bank operations as a proxy for the fight for deregulation. For Daniel Tarullo, this overestimates the significance of differentiating between the two views. Deregulation created a financial system with a shrinking role for traditional deposit-creating (or deposit-taking) banks and increased prominence of ‘shadow banking’ institutions, paving a path to the 2008 crisis. Given that the financial crisis arose out of risks that had little to do with traditional banking, for Tarullo, the importance of picking the right view of how traditional banks work is exaggerated. Instead, for Tarullo, managing risks in the financial system requires a stronger regulatory framework for non-banking financial entities, even if that does not exactly mimic traditional banking regulation.
We might differ with Tarullo however, if we believe that views about banking are embedded within larger frameworks about the relationship between finance and economic activity. As Christine Desan argues, the intermediation view about banks bears an emphatic connection with the idea that economic activity is the outcome of individual initiative: saving and borrowing. Taken as truth, this casts a shadow over our understanding of the history of economic development and the role of state institutions. Thus, Desan places the debate about banking operations within a set of discourses that constitute the neoclassical veridiction for the State’s role in the economy. We are led to mistakenly believe what the drivers of economic development were, and thus to forget that it was creation of novel forms of credit, not the accumulation of existing funds, that prompted capitalist growth.
Like Desan, Hockett and Omarova take to task assumptions about the essentially ‘private’ nature of banking activity. Flipping the intermediation account on its head, they argue that banks intermediate between us (as the sovereign people) and our ‘selves’ (as private actors). In their account, when a bank creates money by issuing credit, this is a way of transforming our private credit-worthiness into a public form, backed ultimately by the ‘full faith and credit’ of the polity. It is this ‘full faith and credit’ that banks are capable of putting into circulation – making them franchisees of the ultimate asset of the sovereign state.
As I noted at the start, at the heart of this Roundtable prompt was the question of what is at stake when we investigate what is it that banks do? Here, I might simply paraphrase from Christine Desan’s contribution addressing what her colleagues understood the stakes to be – matters of disciplinary accuracy, normative questions of regulation and macroeconomic policy, the relationship of public power to private economic action, or of “expositional fit.” To this mix, she adds her own – our understanding of the process of economic development, and thus she concludes that “the stakes could not be… bigger.” I am tempted to agree with her. How we interpret the world of banking appears linked to how we understand some fundamental aspects of production and distribution in society. The distributional stakes remained somewhat hidden from view in this Roundtable, but they lurk under the surface. If banks are the “means of money creation,” then the better account of banking will offer the better account of inequality. We will still have to answer whether a firmer understanding of banking fulfills the great task of philosophy – which is to change the world (for the better)!
 I owe this phrase to Christine Desan – it emerged during our conversations over the last two years.