The word stem crypto within the term cryptocurrency might be seen as surrogate for cryptography, but could also have emerged from the cypherpunk movement, who identified “anonymous cash and other untraceable payment systems” (De Filippi & Wright, 2018, p. 19) as enabling feature within a crypto-anarchy (Ludlow, 2001, p. 4). Bitcoin's mission of leveraging “cryptographic proof instead of trust” (Nakamoto, 2008, p. 1) resonates with the above. The exact protocol specifications of Bitcoin and its descendants are summarised in Scheuermann and Tschorsch (2016). Cryptography enters its architecture in various ways. A few examples are the integrity of, and consensus on a joint transaction history as well as the authorisation setup for sending tokens. However, the use of the surrogate crypto for Bitcoin is slightly arbitrary in the sense that earlier attempts at creating digital currencies (compare e.g. Chaum, 1988) relied heavily on cryptographic techniques as well. Nevertheless, it might seem justified by the fact that cryptography plays a far more central role for Bitcoin than it does for national currencies.

Loosely speaking, the modern fiat monetary system consists of physical and digital credits—issued by state central banks, state treasuries, and private commercial banks—which circulate under a legal system that guarantees their redemption. The number of credits expands through issuance, after which they can be transferred in the course of exchange among those who use them, before being retired when they are returned to the issuers. This composite system of expandable-contractable credits is what we refer to as ‘money’ in everyday parlance. In this context, the term cryptocurrency is controversial, because—from its inception—the name has simply assumed that the tokens are money tokens. The controversy is amplified by the fact that enthusiasts sometimes use the term performatively to make the normative point that crypto tokens ‘should be money’, or—alternatively—to deny that what we currently call ‘money’ is in fact money. One strategy to negotiate these language politics is to initially strip the money assumption from the tokens by giving them the generic name crypto-tokens, and then listing their uncontroversial characteristics to compare them with fiat credits.

Tokens of early cryptocurrencies are data objects created through accounting, much like the act of typing out the number ‘1’ creates the mental image of a ‘thing’. This is what is referred to as a ‘token’, but they are ‘blank tokens’. An example of a blank token in the physical world might be a clear plastic token with no inscription or rights attached to it. Bitcoin tokens, similarly, are empty signifiers, somewhat like the digital equivalent of blank physical tokens, but with strict supply limits

Note that the notion of a “blank token” refers here to economic intuition rather than technical implementation. In Bitcoin and its descendents no “coins” exist, but only transaction outputs that are transferable and arbitrarily divisible.

Most ‘purchases’ conducted with bitcoin tokens, however, take the form of countertrade. The token, priced in fiat currency, is compared to a good or service, priced in fiat currency, and from this comparison of two fiat currency prices emerges an exchange ratio between the token and the good or service. This is the conceptual equivalent of superimposing a pair of two-way fiat currency transactions over each other and cancelling out the money flows, giving the residual appearance of the crypto-token being used as ‘money’ to ‘pay’ for a good or service.

Nevertheless, Bitcoin is used primarily for speculation (Baur, 2018)—buying the token with fiat currency with an intention to resell it for fiat currency—rather than using it to countertrade (‘pay’) for goods and services. This speculation (compare, among others, Yermack, 2015; Glaser et al., 2014; or Cheah, 2015) drives volatility in the fiat currency price of tokens, which—when analysed through the lens of the conventional ‘functions of money’ paradigm favoured by economic textbooks (money as a medium-of-exchange, a store-of-value and a unit-of-account), poses problems for the ‘moneyness’ of the tokens. Not only are they not widely accepted in exchange for goods and services, but they are not widely used to price things, and attempts to provide prices are unintuitive

Usual consumer goods priced in Bitcoin, for example, are represented by tiny decimal numbers.

A useful classification of projects from a technical standpoint involves rights for writing and reading transaction records. Peters et al. (2016) introduced a popular categorisation that can be used to classify the underlying infrastructure of cryptocurrency systems along the dimension “public” vs. “private” and “permissioned” vs. “permissionless”. In public-permissionless systems every participant in the network (node) can read transactions and write others to the ledger. For public-permissioned systems, only authorised nodes can write. In private permissioned systems, finally, even reading is restricted to authorised nodes. The more “private” and “permissioned” in its underlying infrastructure a system is, the further it is from the cypherpunk vision.

An example of a recent development trend holding true to the aim of replacing trust by cryptographic proof found in archetypal cryptocurrencies (compare Nakamoto, 2008; and Genkin et al., 2018) are so-called privacy-preserving cryptocurrencies or ‘privacy coins’ (e.g., Zcash, n.d.; Monero, n.d.). They are closely related to archetypal cryptocurrencies and replicate their public-permissionless setup of rights to read and write. As “alternative cryptocurrencies designed with the goal of providing stronger privacy guarantees than Bitcoin” (Genkin et al., 2018) they even increase the use of cryptography to ensure anonymity. As a consequence of their focus on privacy, however, they are leading to rising concerns with respect to anti-money-laundering and law enforcement (compare Tziakouris, 2020; or Ferrari, 2020).

The broad trajectory in recent years, however, has been to decrease the centrality of cryptography in the respective implementations. Even permissioned payment systems run by corporations but still called cryptocurrencies entered the stage.

Compare e.g. Diem (n.d.) and their reception in the press e.g. New York Times (Popper & Isaac, 2020).

While many novel cryptocurrencies are far from the crypto-anarchist roots of archetypal token designs, the general idea of the replacement of trust in institutions or their internal governance mechanisms by cryptography still plays a role in all cryptocurrency designs. However, given that even fiat bank payments use cryptography for security, mere reliance on cryptography for security should not enter a definition of cryptocurrencies.

We would have liked to rely on the unifying element of blockchain-based technology (which supposedly amalgamates all the cryptographic tools of a cryptocurrency) here. However, noting that this term is similarly unclear and vague as the term to define, we abstained from that step.

Early cryptocurrencies had the declared intent of creating ‘digital cash’ or currency (see section 1.1.), but the proliferation of crypto token forms have destabilised how this is conceptualised. Not all development strands feature the objective of proposing general purpose monetary tokens.

First-layer tokens (e.g. Ether) that underlie smart contract platforms

A summary of the research around smart contract platforms is given in Macrinici et al. (2018) while Bartoletti et al. (2017) and Alharby et al. (2017) review different platforms. While generally similar to cryptocurrency systems, their tokens are part of the security setup and used as medium-of-exchange between smart contracts.

However, also ‘general purpose’ tokens are marked by changes. A response to the inherent instability in prices of archetypal cryptocurrency was the advent of ‘stablecoins’, which try to solve the issue of high volatility in purchasing power of Bitcoin and its descendants (Pernice, 2019). Stablecoins are tethered or pegged to fiat currencies, or ‘backed’ in some way with assets that have fiat currency prices. They are thus no longer ‘blank’ empty signifiers, and contain some reference point that is easier to estimate and communicate. There are very different types of stablecoins, and recently several frameworks have tried to unify and abstract existing stabilisation techniques (e.g., Bullmann et al., 2019; Pernice et al., 2019; Moin et al., 2020; Sidorenko, 2019; Clark et al., 2020). A national currency can be ‘tokenized’ by issuing a digital promise for it on a blockchain system, and such tokenised funds might indeed be categorised as a “new form of electronic money” (Blandin et al., 2019) falling under the respective regulations for e-money, anti money laundering and counter terrorist financing regulations. This might ensure “moneyness” at least from a legal standpoint. With more complex stablecoin designs the legal case is not always clear, but from an economic standpoint their stability in purchasing power might contribute to an increase in their adoption as money in the future. Stablecoins, for now however, haven’t seen mainstream adoption in retail markets yet (Bullmann et al., 2019).