Historian Harold James of Princeton University, known for his scholarly writings on the gold exchange standard and on the euro, has turned his attention to Bitcoin in a recent Project Syndicate commentary on “The Bitcoin Threat.” His commentary labors under a surprising number of misconceptions about Bitcoin and the history of privately issued currency. If even a reputable academic historian falls prey to these misconceptions, they are likely to be widespread. Scrutinizing them may then be of wider interest.
After noting some of the optimistic claims made on behalf of cryptocurrencies and the blockchain technology underlying them, James cautions us:
But others are rightly suspicious that this new technology might be manipulated or abused. Money is part of the social fabric. For most of the history of human civilization, it has provided a basis for trust between people and governments, and between individuals through exchange. It has almost always been an expression of sovereignty as well, and private currencies have been very rare.
To say that government-issued currencies have “provided a basis for trust,” and to imply private currencies have not, is a curious summary of centuries of monetary history. Anyone familiar with the long history of debasements by ancient and medieval government mints, or with the history of fiat money inflations by modern government central banks, knows that governments have often been untrustworthy issuers. Sovereigns have frequently abused rather than rewarded trust in their currencies. (To his credit, James does later observe that “bad states produce bad money.”) Indeed a key service that attracted medieval merchants to private bankers was their more trustworthy payment alternative to the variously debased government-issued coins, namely a ledger-based system where transferable account balances were denominated in units of unchanging silver content. Historians later called these stable private accounting units “ghost monies” because they were not embodied in any of the debased contemporary coins.
James’ statement that “private currencies have been very rare” is simply untrue. It is a surprising misconception for a financial historian to hold. Private silver and gold coins were historically rare, it is true, because governments have legally suppressed private mints to give their own mints monopoly privileges. But during the 18th and 19th centuries, redeemable paper currency became more popular than coins in modern economies, and the majority of paper currency in circulation in most countries consisted of privately issued banknotes. Kurt Schuler and Will McBride report that more than sixty countries have had periods of competitive private note-issue, so it was hardly a “very rare” experience.
Banknotes are a banking product, and banking is normally a private business, so it should not be surprising that the very first known banknotes were introduced by businesses and not by a government. Schuler and McBride write that “the first true circulating notes were issued” around the year 995 “by private bankers in the city of Chengdu” in China. Similarly, “In Europe, the first true circulating notes were issued in 1661 by Stockholms Banco, a private bank chartered by the crown.”
In the United States before the Civil War, the vast majority of paper currency was issued by private state-chartered banks. The only governmental or quasi-governmental notes were those issued by the short-lived first and second Banks of the United States (whose shares were 80% in private hands) and by state-government-owned commercial banks in a few frontier states. After the Civil War, private banks with federal charters (the “National Banks”) continued to issue notes into the 1930s. In the United Kingdom, all banknote issuers including the Bank of England were private before the First World War. Even if we were to consider the BOE “non-private” after it gained special privileges in the Bank Charter Act of 1844, other private banking companies issued 43 percent of the notes in circulation in 1851, to pick a mid-century year. Canada’s banknotes were entirely private before the provinces (later The Dominion) began issuing small notes in 1866, and there private banknotes continued in circulation until the 1940s.
When it comes to the operations of Bitcoin, James exhibits additional misconceptions. He comments that “Bitcoin looks like a twenty-first-century version of gold, and its creators have even embraced that analogy.” I discussed here recently what the Bitcoin system has in common with a gold standard, and how it differs in important ways. James’ statement that Bitcoin’s “creators have even embraced that analogy” seems to refer to the term “mining” being used metaphorically to refer to the operation of Bitcoin validation nodes. But in writing that Bitcoin “is produced — or ‘mined’ — through effort,” James appears to have been misled by the term. Bitcoin “mining” validates payments. Unlike gold mining, growth in the number of computers “mining” bitcoin does not increase the rate at which the total stock of bitcoin grows. It only adds more competition to be the mining node that receives the reward in new coins for processing the next transaction block (by being first to solve a mathematical guessing problem, the difficulty of which is endogenously adjusted to keep the expected solution time at ten minutes). New coins are awarded at a rate that is predetermined by the source code. Thus, a new miner who adds a computer to the system “produces” expected new bitcoins for himself, but not for the system as a whole.
The most dramatic of James’ ill-founded claims come in the commentary’s penultimate paragraph — which is hard to read as anything but rather wild fear-mongering:
And yet we have already reached the point where a Bitcoin crash could have serious global implications. Financial institutions’ current exposure to the cryptocurrency is unclear, and probably would not be fully revealed until after a financial disaster. It is eerily reminiscent of 2007 and 2008, when no one really knew where the exposure to subprime-mortgage debt ultimately lay. Until the crash, it was anyone’s guess which institutions might be insolvent.
James cites no evidence of financial institutions’ exposure to cryptocurrencies. Unlike mortgages in 2007, known commercial and investment bank balance sheets indicate no significant holdings of cryptocurrency assets by the institutions. Asked about cryptocurrencies, European Central Bank chief Mario Draghi commented on 5 February 2018 that bank exposure was not evident: “Let me first say that we are not observing a systemically relevant holding of digital currencies by supervised institutions — by banks, in other words.” As he went on to add, banks do not hold them precisely because cryptocurrencies are indeed very risky investments. A Forbes blogger in December, imagining bank losses from cryptos, was to his credit willing to acknowledge: “Of course, this is a doomsday scenario and there’s no evidence that big banks have garnered large positions in Bitcoin or other currencies — yet.”
The only apparent (albeit minuscule) exposure of banks, one mentioned in recent news reports, is via credit-card customers who charge such large cryptocurrency purchases to their cards that a plunge in crypto prices might make them personally bankrupt and force them to default on card payments. No problems from credit-card losses of this kind have yet been reported, however, and the card-issuing banks are fully able to price or quantitatively limit such a risk. Coindesk reports that JPMorgan Chase, Bank of America, Citi, and Capital One are now in fact disallowing cryptocurrency purchases by their credit-card customers. Of course the banks already cap the size of their customers’ credit lines to limit default risk, and charge additional fees when they allow customers to take (limited) cash advances.
A bank that allows its customers to use a debit card or other form of deposit transfer to buy cryptocurrencies, it should be noted, is not extending credit to those customers and is not exposing the bank to any risk of credit losses from crypto price volatility.
In 2007, by contrast, regulators looking at audited balance sheets knew which banks were holding how many billions of dollars in mortgages and mortgage-backed securities (MBS). (Granted, the regulators were unaware ex ante of how risky the mortgages and MBS were.) Total MBS outstanding in 2007 and 2008 were over $9 trillion, by the way. The total volume of cryptocurrencies outstanding today is under $400 billion, less than 1/20th of $9 trillion, and cryptocurrency purchases on outstanding bank card balances must be a very tiny fraction of that. Again, there’s no evidence of banks or investment banks holding crypto positions. So today’s situation is hardly “eerily reminiscent” of 2007-08.
James’ phrase “financial institutions” covers more than banks, of course. And there are a small number of speculative financial institutions buying cryptocurrencies for their customers, namely specialized hedge funds and proprietary trading firms. Pantera Capital offers a leading cryptocurrency investment fund. The CEO of Pantera notes that the crypto market is remarkable for the nearly complete absence of institutional investors: “It’s a half-a-trillion-dollar asset class that nobody owns. … And bitcoin is still so underowned by institutional investors that it trades at its own beat.” At the mid-December BTC price peak, the firm reportedly had about $2 billion under management. Half of that value has been lost, but without any apparent spillover effects. Hedge fund shareholders are high-net-worth individuals who can afford speculative losses to parts of their portfolios. Contrary to James’ fear that “a Bitcoin crash could have serious global implications,” the halving of Bitcoin’s value over the last two months shows no worrisome spillovers to the financial system more generally. It provides no rationale for restricting the public’s right to buy or hold or use cryptocurrencies.