1 Introduction/rejoinder

Let me begin by thanking Yuri Biondi and Accounting, Economics and Law: A Convivium for hosting this book review symposium. It is a privilege to have my book reviewed by this distinguished roster of experts. Since the book’s publication I have had some time to reflect on its strengths and weaknesses. Unsurprisingly, the reviewers in this issue have identified a number of the book’s more glaring shortcomings. But it relieves me to say that I don’t think the book’s key arguments have (yet) sustained any mortal wounds, even if solid blows have been landed.

The basic thesis of The Money Problem is that financial instability is, and always has been, mostly a problem of private sector money creation. This is an old-fashioned view. I think it’s pretty clear that most experts today don’t subscribe to it. I continue to think it essential that financial instability be understood explicitly in terms of money creation. This means treating financial stability policy as an aspect of monetary system design—an institutional design project that implicates not just stability but also questions of monetary control and seigniorage, or fiscal revenue from money creation. I see these topics as components of a single, integrated institutional design challenge.

The book lays out a blueprint for monetary and banking reform. The proposed reform would follow through on the logic of U.S. banking regulation as it has existed since the New Deal era. While in this sense the blueprint is (conceptually) conservative, it would require some very big alterations to existing arrangements. Rather than summarizing the blueprint in any detail, let me just touch on a few of its salient features. The proposal would restrict entry into “money” creation. Money creation, for this purpose, means issuing large quantities of short-term or demandable debt (denominated in the standard unit of account) that is continuously rolled over. Money creation is thus associated with a specific funding model. This funding model would be limited to licensed banks; the banking system and the government itself would be the exclusive issuers of money. In effect, banks would be engaged in a public-private partnership—a joint venture with the state for the issuance and circulation of money. The monetary authority would control the quantity of money issued by the banking system. All money would be fully sovereign and nondefaultable, backed by the government. Banks would pay risk-based fees to the state, which would flow to fiscal revenue (seigniorage). I argue that this basic structure would greatly enhance financial stability while also improving macroeconomic management and enabling the public to recapture economic rents currently captured by elements of the financial sector.

Blair (2018) homes in on one of the book’s key arguments: my claim that, insofar as financial instability policy is about preventing acute macroeconomic disasters, it should concern itself mostly with panics, or sudden widespread redemptions of the financial sector’s short-term and demandable debt (or “money-claims”). Why did the U.S. economy shed about seven million jobs in a matter of months in late 2008 and early 2009? Why did economic output plummet? Blair thinks I have not given enough attention to the role of collapsed asset price bubbles—particularly debt-fueled bubbles—in harming the real economy. I do treat this issue in chapter 4 of the book, but Blair is not wholly convinced.

I acknowledge that my arguments on this score are not definitive, but the debt-fueled bubble explanation is far from definitive either. For example, while Mian and Sufi (2014) show that from 2007 to 2009 spending fell more in U.S. counties with larger declines in housing net worth than in counties with smaller declines, and that the decline in jobs catering to local demand was larger in counties with larger declines in housing net worth, they do not estimate the impact of this channel on aggregate U.S. employment. It has always seemed telling to me that the U.S. housing collapse started in earnest in the first quarter of 2007 and was already two-thirds played out when Lehman Brothers failed in September 2008, yet prior to Lehman the U.S. economy was in only a mild, garden-variety recession. With the onset of the severe panic, the character of the recession changed drastically. Industrial production and personal consumption expenditure fell more in September 2008 than they had in the previous nine months combined. The freefall continued; the decline in these measures was three times greater over the subsequent nine months (the period in which money market distress was most acute) than it had been in the nine months preceding the Lehman bankruptcy. I show in the book that the timing of the U.S. macroeconomic collapse matches up extremely well with panic-related financing constraints but not with declines in housing wealth. Admittedly this is not a slam dunk case, but it is highly suggestive; it should at least shift the burden of persuasion. I do not deny that collapsed bubbles per se are bad for the real economy, but it seems to me that Friedman and Schwartz’s conclusion still holds up: “Banking panics,” they wrote in their seminal (1963) study of U.S. monetary history, “have occurred only during severe contractions and have greatly intensified such contractions, if indeed they have not been the primary factor converting what would otherwise have been mild contractions into severe ones.”

Moutot (2018) addresses this issue from a somewhat different angle. He too thinks I ascribe too much significance to panics as a source of severe recessions. He uses Japan’s lost decade(s) and Europe’s postcrisis stagnation as illustrations; both of these episodes, he rightly says, have to be understood through different or at least more general frameworks. Moutot also suggests that the “cap and trade” approach to monetary policy that I outline the book is tantamount to using monetary aggregates as a direct tool of monetary policy, an approach largely discredited decades ago. And he raises Fintech as a challenge to any effort to confine money creation to chartered banks. Fintech, he points out, can be a source of regulatory arbitrage; for example, would Bitcoin have to be banned? Finally, Moutot expresses legitimate concerns about the transition costs associated with requiring nonbank financial institutions to adjust their funding models away from money-claim funding.

Contrary to Moutot’s reading, I did not mean to suggest that panics “constitute the actual and single cause of severe recessions.” I should have been clearer on this point. No doubt bad monetary policy or sovereign debt crises, to take two examples, can cause or amplify severe recessions. What the book argues is that panics are far and away the biggest danger the financial sector poses to the real economy and that panic avoidance should therefore be at the core of financial stability policy. This is a narrower claim and, as I noted above, I still think the weight of the evidence points in this direction. Moutot’s Fintech point is well taken; the prospect of regulatory arbitrage cannot be avoided if money-claim issuance is to be restricted by regulation, and difficult line-drawing problems are inescapable. (My approach to entry restriction as outlined in chapter 9 would not outlaw Bitcoin because it is not dollar-denominated, but Moutot’s broader point stands.) And I do think transition would need to be gradual in order to avoid undue disruption. In retrospect I might have said more about this in chapter 9’s “Getting There from Here” section.

Moe (2018) likewise delves into the conceptual and boundary issues that arise in defining what claims constitute “money” and in restricting their issuance. Moe’s erudite review brings to my attention a fascinating (2014) document from the European Banking Authority (EBA), which I had not previously seen, in which the EBA asked the European Commission for definitional guidance: “The key terms ‘deposits’, ‘other repayable funds’, ‘grant credits’, ‘from the public’ are not defined in the [capital requirements regulation],” the EBA noted, leading to uncertainty as to “the entities to which the requirement to obtain a banking license applies.” Moe sees a possible “category mistake” in treating short-term debt as “money proper.” And he compares and contrasts my proposed regulatory design to alternative reform proposals, including Tobin (1987) “deposited currency” proposal as well as more recent “central bank digital currency” (CBDC) proposals.

The definitional or boundary issue is indeed critical to my overall project. As I noted above in my comment on Moutot’s review, there is no getting away from line drawing. This of course is a standard regulatory problem; securities regulation must define security, investment company regulation must define investment company, proprietary trading regulation must define proprietary trading. In all of these areas the unavoidable starting point is to define in functional terms what type of instrument or activity is being regulated. There are always gray areas and boundary problems, but I don’t think they present an insuperable obstacle to regulation. The question is whether “money-claims” are different in some essential way, and I’m not convinced they are. As I show in the book, historical attempts to define “deposit” in legal-regulatory terms have invariably been formalistic; a functional approach has never been tried, at least not in Anglo-American banking law. As for Moe’s comparison of The Money Problem’s proposed regulatory design to Tobin’s deposited currency and CBDCs, he is spot on. These policy ideas share similar motivations. I have recently come around to the view that allowing (though not compelling) everyone to hold accounts at the central bank—“central banking for all”—could be a nice way of achieving, by alternative means, some of the policy aims of The Money Problem.

Biondi (2018)’s contribution to this issue—which is not a review of The Money Problem but rather a deep scholarly analysis of money and banking—deserves attention both for its conceptual apparatus and for its linkages to two centuries’ worth of money and banking literature. Biondi’s paper is a rich mine of citations to money-and-banking classics, including references to several works and passages I had not previously read or noticed. (For example, I was unaware that Sayers defined liquidity as the combination of at will and at par redemption; this precedent would have helped me a great deal in chapter 1 of the book.) Like me, Biondi finds much to be gained from viewing banks as money issuers rather than purely as financial intermediaries. This “money” view affords a systemic perspective on banking, highlighting its reliance on horizontal coordination and interconnection. Biondi sees the banking system fulfilling a threefold function of ledger keeping, treasury management, and manufacturing of money. Rather than offering a summary or analysis of his framework, I encourage you to read it.

Even now, a decade after the Global Financial Crisis and the subsequent waves of financial reforms, dissatisfaction with existing monetary and banking arrangements remains widespread both in the United States and in other jurisdictions. The recent Vollgeld referendum in Switzerland, which would have banned fractional reserve banking, is evidence of this continuing unease. The Money Problem presents an alternative vision, but it isn’t the only one available. The contributors to this issue have varying diagnoses and prescriptions. But we all seem to agree that the project of designing money-and-banking institutions that serve the public interest remains unfinished.