At the root of much of the debate about Bitcoin is confusion about the nature of money.
In the textbook treatment, money in an economy serves as (i) a medium of exchange, (ii) a unit of account, and (iii) a store of value. Or, in everyday language, money is the item that we offer or receive for the purchase or sale of goods, services or assets—it is simply a device that facilitates transactions in a market economy. The confusion arises because the physical properties of money are practically arbitrary—paper bank notes, gold coins, digital tokens or stone disks could all serve the purpose.
Instead, the value of money derives from its usefulness. Therefore, digital currencies like Bitcoin could prove to have value, if they can demonstrate their usefulness relative to existing forms of money.
Demand for money comes from two sources: transaction demand and portfolio demand (Exhibit 1). Economists have a number of theories for the transaction (or liquidity) demand for money, but they all boil down to the same thing: we cannot easily exchange (non-monetary) financial assets (“assets”) for goods and services, so we tend to hold some amount of monetary assets (“money”) to facilitate consumption. Transaction demand for money declines with expected returns on assets (the opportunity cost of holding money), increases with transaction costs (like brokerage fees), and increases with income (because more transactions require more money). Portfolio demand for money also declines with expected returns on assets, and one could argue this is why some precious metals (such as gold) are more like money than other assets. Finally, portfolio demand for money increases with uncertainty about asset returns and increases with wealth (because more wealth means larger money balances).
Nominal exchange rates are the relative price of two forms of money, and their fluctuations result from these same demand drivers. An exchange rate appreciates because of higher transaction demand—e.g., due to export competitiveness or an increase in cross-border M&A—or because of higher portfolio demand—e.g., investors expect relatively higher returns or relatively lower risk on assets denominated in that currency. Although we do not often consider this factor for major exchange rates, a currency could also appreciate because it provides lower asset-to-money transaction costs, and therefore becomes more useful for transaction purposes (a possible selling point of cryptocurrencies).
When Money Malfunctions
To a modern American observer, cryptocurrencies can seem like a solution in search of a problem. In recent decades the US Dollar has served its purpose relatively well: consumer price inflation has averaged 2.1% over the last 30 years, and the real trade-weighted exchange rate is about 3% above its average of the same period. It is true that the US financial system wobbled during the financial crisis, but the underlying problem was not the currency per se: the Dollar actually appreciated 16% between June 2008 and March 2009. The Dollar accounts for about 65% of global foreign exchange reserves and is the denominate currency in global trade—about 30% of global trade flows excluding the US are invoiced in USD. In other words, transaction and portfolio demand for the Dollar are very high.
The same cannot be said for many other types of money. In certain parts of the world, and at points throughout history, multiple currencies have circulated in the same economic area. In academic research this is called “currency substitution”, but it is more commonly known today as Dollarization. Many countries with a history of currency instability and/or underdeveloped financial markets use the US Dollar for the traditional functions of money—as a medium of exchange, unit of account and store of value. In sub-Saharan Africa, for example, many currencies have been debased by high inflation and the mismanagement of money supply. As a result, Dollarization is widespread: foreign currencies account for more than 90% of deposits and loans in the DR Congo, and Zimbabwe demonetized its own currency in 2015 (a variety of foreign currencies now circulate). The US Dollar is the official currency of Ecuador and a variety of smaller nations, where no separate legal tender circulates. The Peruvian economy remains significantly Dollarized today despite stable inflation over the last decade—a legacy of its experience with hyperinflation in the late 1980s. In these circumstances, it is natural to see how alternative—possibly digital—anchor currencies could be useful as well.
Government regulation can also restrict the use of outside currencies in various ways.
For example, countries can limit the amount of foreign exchange that can be held in bank deposits (e.g., Mexico) or by certain types of investors (e.g., pension funds in Poland). They can also tax foreign exchange transactions or maintain multiple official exchange rates, and capital controls may limit the degree to which residents can convert domestic money into foreign exchange. Prominent recent examples include: (i) the prohibition on purchases of foreign assets by Greek banks, (ii) limits on currency conversion in Cyprus, and (iii) the use of various quotas for overseas investment by China. In situations such as these, where residents are unable to accumulate traditional foreign exchange, a decentralized currency such as Bitcoin could attract significant demand, especially if it is not (yet) legally recognized as a currency.
Lastly, while Bitcoin and other cryptocurrencies employ innovative technologies, there is nothing particularly unusual about new types of money or monetary systems. Probably the best recent example was the introduction of the Euro in 1999. The Euro began as a fixed exchange rate system between 14 EU member states, but is now the official currency of 19 EU members plus a handful of smaller European nations, and is the second most common currency held in foreign exchange reserves and used in global trade—partly displacing the US Dollar. Some economists make the case that the IMF’s Special Drawing Rights—in a redesigned format—could also play a role as an international currency in the future. In theory at least, there is no reason why cryptocurrencies could not do the same.
We see some evidence that the demand for cryptocurrencies could be related to dissatisfaction with current monetary systems. For instance, a Google Trends search shows that the highest search intensity for “Bitcoin” over the last five years (scaled to overall search volumes in the country) came from Nigeria, South Africa and Ghana, all countries with currency instability and/or restrictions on the use of foreign exchange—places where alternative forms of money might see natural demand. In addition, Bitcoin exchange volumes in China rose sharply following the tightening of capital controls in 2016. They subsequently collapsed when cryptocurrencies were more stringently regulated last year (Exhibit 2).
Other data look more consistent with a classic speculative bubble. First, Bitcoin exchange volumes are now dominated by investors in Korea and Japan—countries with no recent history of monetary instability and/or unmet portfolio diversification needs (Exhibit 3). Second, cryptocurrency prices are correlated with Google search volumes (Exhibit 4). This may point to significant retail investor participation in these markets.
Third, until the launch of Bitcoin futures, it was generally not possible to short the market—a common feature of speculative bubbles.
A remaining question is what to make of the many types of digital currencies on the market. According to the website coinmarketcap.com, there are currently more than 1,300 different cryptocurrencies available today (including both coins and tokens). As discussed above, the value of money comes from transaction demand and portfolio demand. However, the value of a particular type of money is also enhanced by network externalities. The US Dollar derives some of its value from the fact that it is universally accepted as payment in the United States, and widely accepted as payment throughout the world. For a specific cryptocurrency to succeed, it may require support from network externalities—i.e., a “critical mass”—but this may be challenging given the large number of coins available.
A Case for Crypto?
In practice, Bitcoin and other digital currencies face significant practical hurdles to their adoption as outside forms of money, and many of their possible benefits come with significant drawbacks—several of which were highlighted by our colleagues in an earlier report (Top of Mind: All about Bitcoin, March 11, 2014).
First, some features of cryptocurrencies that might make them competitive with alternative stores of value are also features that are likely to attract government scrutiny.
In particular, the anonymity of many cryptocurrencies makes them a useful medium of exchange for criminal activities, including tax avoidance and the circumvention of capital controls. As such, it would be surprising if continued growth in their popularity did not eventually attract greater regulation and law enforcement action by government.
Second, the fact that cryptocurrencies function without central banks may make them valuable as inflation hedges or stores of value, but also makes them vulnerable to demand-driven fluctuations in price. Such volatility makes them poorly suited as a substitute for money generally—which is why most nations eventually abandoned the gold standard in favor of fiat currencies that can more easily stabilize the purchasing power of money by making the necessary supply adjustments in response to changes in money demand. The recent fluctuations in Bitcoin and its relatives suggest they are much too volatile to serve as money (Exhibit 5). Volatility would likely need to come down dramatically (either naturally or through the widespread adoption of cryptocurrencies designed to better stabilize purchasing power via supply adjustments) before we see broader adoption.
We should also stress that, as money, cryptocurrencies should have low expected returns in the long run, despite their high returns recently. Our working assumption is that long-run cryptocurrency returns should be equal to (or slightly below) growth in global real output—a number in the low single digits.11 Thus, digital currencies should be thought of as low/zero return or hedge-like assets, akin to gold or certain other metals.
This could still mean that prices increase at a faster rate as the technology is adopted—an analogy might be the value of a biotech company that invents a drug, which eventually becomes a generic—but there is an inherent contradiction between cryptocurrencies as high return assets on the one hand, and stores of value relative to goods and services on the other.
So could Bitcoin succeed as a form of money? In theory, yes, if it proves to be more useful than the alternatives—in terms of facilitating transactions at a lower cost and/or providing better risk-adjusted returns for portfolios. In practice, however, these gains look small, at least in developed market economies. Transaction costs are relatively low, exchange rates and price inflation are broadly stable, precious metals can be used for portfolio diversification, and governments place few restrictions on holding foreign currency or foreign assets. That said, the widespread use of the Dollar outside the US—and full Dollarization in some countries—suggests there is already demand for an internationally accepted medium of exchange and store of value. In those countries and corners of the financial system where the traditional services of money are inadequately supplied, Bitcoin (and cryptocurrencies more generally) may offer viable alternatives.
Zach Pandl - Goldman Sachs
Charles P. Himmelberg - Goldman Sachs