Token Taxonomy & Properties of Tokens

While it is technically possible to represent any asset of the existing econo- my as a cryptographic token, we still lack adequate taxonomy, and adequate legal framework that understands the full scope and potential of this new substrate with which we can issue any type of asset and access right, including completely new asset classes.

Establishing a consistent and reliable taxonomy for token properties, as well as classic cation models, is imperative to laying a foundation from which developers, policymakers, and investors can make more sense of how to design, apply, or regulate tokens.

Regulatory authorities all over the world are catching up to understand the full potential and implications of this new technology. At such an early stage, however, it is hard for anyone to understand what is truly new, from existing asset classes represented by a new derivative. Some tokens might represent completely new asset classes, like native protocol tokens that very o en have hybrid functions, which are not easy to classify, from tokens that might very simply represent assets of the existing economy that are easily classified, understood from a business logic perspective, and therefore also regulated. An example of assets that are easy to classify or regulate are tokenized securities and other assets.

We are still in the very early stages of exploring different roles and types of tokens. Many of the terminologies we use today will adapt to the realities of emerging use cases and should be considered as temporary. With every new blockchain and every new token application, we will collectively learn by trial and error about possible use cases of cryptographic tokens, and resulting classical cations thereof. The taxonomy presented here intends to give a broad overview of the different properties and types of tokens but is by far not complete. It is a big picture of the most important economic, technical, and regulatory questions tied to such taxonomy.

The following chapters intend to give different perspectives on how such classification could be thought of, without any claim to completeness or exclusiveness. The focus will only be on the different properties a token can have. Classifying the properties of a token is necessary for modelling tokens (development perspective) and evaluating tokens (investor perspective). A classical cation and taxonomy of the token itself would be legal, business, economics, and social sciences questions. A legal taxonomy is furthermore subject to a specific jurisdiction, and would be far beyond the scope of this book.

Properties of Tokens

Identifying different properties of a token can be used as a first step to finetune a future classification framework and also used for modelling techniques when developing a token (token engineering). Morphological analysis as introduced by Zwicky can be used to explore all the possible solutions to a multi-dimensional, non-quantified complex problem. This method is especially helpful for the structuring of questions in a first approach in a heuristic way. I would, therefore, like to introduce the most important perspectives from which we can deduce the properties of a token: (I) Technical perspective; (II) Rights perspective; (III) Fungibility perspective; (IV) Transferability perspective; (V) Durability perspective; (VI) Regulatory perspective; (VII) Incentive perspective; (VIII) Supply perspective; and (IX) Token ow perspective.

Technical Perspective:
From a technical perspective, tokens can be implemented on different layers of the technology stack, either as (I) “protocol tokens,” which are part of a crypto-economic incentive mechanism, or as (II) “second-layer tokens,” like application tokens or tokens created on a sidechain. Sidechains are separate blockchains, compatible with the mainchain, and have been used to resolve scalability issues in Bitcoin (read more on Sidechains: Annex – Scalability Solutions). Protocol tokens, also referred to as intrinsic, native, or built-in tokens, have a very clear role in a blockchain: to keep the network safe from attack by acting as block validation incentives (miner rewards), and for transaction spam prevention. The token, in this case, is an integral part of the incentive mechanism to make a decentralized protocol work (read more: Part 1 – Bitcoin, Blockchain, & other DLTs). Native protocol tokens might furthermore be needed to pay for transaction fees in the network. Examples thereof are Bitcoin, Ether, SIA, etc. These tokens are often referred to as cryptocurrencies or coins. However, I personally find these terms misleading. These native protocol tokens can be seen as the currency of the distributed Internet tribe called Bitcoin Network, Ethereum Network, or Sia Network. Technically speaking, however, they are all tokens. Second-layer tokens, on the other hand, can have any function or property. They can represent anything from a physical good, a digital good, or a right to perform an action in a network or in the real world. They can be issued as an application token managed by the underlying blockchain. In the case of Ethereum, Ether (ETH) is the native protocol token, and ERC-20 and other Ethereum token standards allow the creation of application tokens with a smart contract. Second-layer tokens can also be issued by a sidechain. This is more dominant in the Bitcoin ecosystem. Sidechains that allow the creation of second-layer tokens are, for example, Elements, Liquid, or Rootstock. With a few lines of code, application tokens can be easily programmed to represent any right mentioned above. They interact with a blockchain to manage the state of the tokens. Due to network effects, the value of second-layer tokens is likely to be interdependent with the value of the underlying native blockchain token. An example thereof is the value of ETH (the native Ethereum token), which rose in the ICO bubble of 2016 to 2017 due to the large amount of ETH that was needed to buy app tokens issued through ICOs. Multi-asset ledgers like Ripple and Stellar allow the creation of multiple tokens on the native blockchain level. Stellar allows anyone to create token contracts with all kinds of variables. On Ripple (XRP), everyone can issue any kind of token on the network, but they are issued as IOUs,32 essentially debt. XRP is considered credit, which is why some call it the “credit network.” To get to use these tokens, others must enable trust to one’s wallet, which means that one transfers debt.

Rights Perspective:
Tokens can represent a right to some underlying economic value, whether digital or physical, long term or temporary. A token can represent (I) a right to an asset I own, which can be a unit of account or a unique good, or (II) limited access rights to an asset someone else owns. The economic dffinition of an asset is a resource which has an economic value and is controlled by an individual or a legal entity or a country. The legal dffinition of an asset is anything which has monetary value attached to it. Ownership right is legal right to possession of a thing, including all usage rights (physical and intellectual). In some countries, ownership is only possible in connection with physical things. Rights of use, or access rights, are contractual rights to use something in possession of someone else. Therefore, a token could represent any asset or resource representing one’s ownership or the right to use said resource. These can be public or private assets, utilities, or services of any kind. However, this is not a binary classification, as many use cases might be more hybrid in nature, like mining rights on a piece of land, which are an access right but also represent a productive asset. Native protocol tokens like Bitcoin (BTC) and Ether (ETH) can be seen as assets, but also represent access rights to the network, as they are needed to pay for network fees. If tokens represent assets, they act as a passive payload managed by a distributed ledger, including all properties, rights, and obligations it has in the system. Asset-backed tokens can be (I) fungible or (II) non-fungible. Fungible tokens represent ownership of any fungible physical goods like at money, silver, petrol, gold, diamonds, shares in a company, or any collateralized debt instrument. They could be compared to commodity money and are therefore sometimes referred to as crypto-commodities. Asset-backed tokens can also be unique and therefore non-fungible. Some refer to them as crypto-goods. Examples would be real estate tokens, crypto-collectibles, or tokens that represent unique pieces of art. Representing such assets with a token makes the asset more easily tradable and divisible, thus creating more liquidity for some assets that might not have been that easily tradable off-chain. Tokens can also represent access rights that are limited in time or in scope of using an asset someone else owns or a service someone else provides. They can provide access to network services, an entry ticket to a concert, a public transport ticket, apartment sharing access, car sharing access, a time slot for a doctor‘s appointment, or membership access to a club, just to name a few examples. They could be used to allow you to start your car, which might have a smart lock, access to alcoholic beverages proving that you are above a certain age, board an airplane, enter your home, vote, cross a border, collect a tax refund, or get a discount, just to name a few examples.

Fungibility Perspective:
In economics, fungibility refers to the interchangeability of each unit of a commodity with other units of the same commodity. Examples thereof could be any durable goods, like precious metals or currencies. Fungible assets have two key properties: (I) Only quantity matters, which means that units of fungible assets of the same kind are indistinguishable. (II) Any amount can be merged or divided into a larger or smaller amount of it, making it indistinguishable from the rest. If you were to lend 10 EUR to someone, for example, it would not matter if that person returns the exact same 10 EUR bill or another one, or various bills and coins that amount to the value of 10 EUR. The same applies to one barrel of crude oil. Flour is another example of a fungible asset, and is also one of the reasons why it was used as a commodity currency in the past. Fungibility is the essential feature of any currency or commodity if it needs to act as a means of exchange, a unit of account, and store of value. Equally, fungible cryptographic tokens can represent any physical or digital assets that are identical to each other and can therefore be easily replaced. They are not unique and are perfectly interchangeable with other tokens of their kind. If two parties have the same amount, they can swap them without losing or gaining anything. Unique tokens, on the other hand, are non-fungible. Examples thereof are ID cards, a token that represents the ownership of a house, car, piece of art, or a gym membership. If you lend a non-fungible token that is transferable to someone, you would expect them to return the same token. This is similar to a book that you lend to a friend. You would expect them to return the same book to you, not another book.The more easily divisible a token is, the more fungible it is. Divisibility refers to the fact that you can send a fraction of the token to someone else. In the real world, many real assets cannot be divided, which makes them less easily tradeable. One advantage of cryptographic tokens is that on a blockchain, tokens representing goods, which were not easily divisible before, can now be fractionalized at lower transaction costs than with established systems. Physical goods that are non-divisible can be first tokenized, and then divided and sold in diferent parts. Fractional tokenized ownership might allow a new array of asset classes, like real estate or art, and make those assets more liquid and fungible. However, there are practical limitations of redeeming a represented asset, for example a piece of art (read more: Part 4 – Asset Tokens & Fractional Ownership). While in theory, there is no limit to making tokens divisible to 100 decimals, it is not economically feasible to do so. The overhead is huge when dealing with trillions of addresses that can and will hold le over “dust.” Dust in this context is referred to as very small amounts of unspent tokens that are o en not worth it to transfer, as the transaction fees mights be higher than the dust is worth. There is a point where the marginal utility of extra divisibility is outweighed by the extra computational e ort (storage and bandwidth). Furthermore, dusting attacks, where miniscule amounts of tokens are sent to random addresses to make them easily traceable, will be more feasible.

Transferability Perspective:
Tokens can be transferable or non-transferable, or have restricted transferability. Unique (non-fungible) tokens can be transferable or non-transferable depending on the use case. A plane ticket might be transferable or non-transferable depending on the type of ticket you bought. A piece of art, or the registration paper of your car, for example, is unique, but transferable. Identity-bound tokens like certi cates or licences are usually non-transferable. A token that allows you pick up your kids from kindergarten is unique, but could have some limited or temporary transferability to allow you to arrange for someone else to pick up your kids by temporarily granting that person pickup rights. While fungible tokens tend to be transferable in most cases, there are also exceptions to the rule.

Durability Perspective:
In economics, durability refers to the ability of a currency to withstand repeated use. This means that the substrate of that currency should not easily vanish, decay, or rot. Metals or durable foods like wheat have high durability, and were therefore o en used as commodity money. The Bitcoin token and similar native blockchain tokens have so far proven to withstand time, being resilient against any type of censorship or network attack. A resilient network is expected to contribute to a “relatively” stable long-term value of the token. If one can correlate network resilience to the value of the network token, the token can be expected to be durable, as it will not cease to exist. As long as the network is robust and used, new tokens will be minted and demand for tokens will increase. Token prices might decrease due to price uctuations, but the token as such will not vanish as long as the network is intact. A network with a weak consensus protocol, on the other hand, might be attacked and manipulated. In this case, token holders could lose their tokens if the state of the network – the ledger – is tampered with.

Regulatory Perspective:
Regulation is a complex topic that could cover a book on it’s own, especially taking into account all 200+ jurisdictions in the world. To simplify matters, at this point it is sfficient to say that regulators need a clear taxonomy of the different types of tokens to understand what they are potentially regulating. In this context, some of these tokens are easy to classify and regulate, as they re ect known phenomena, while other types of tokens might be much harder to classify and regulate, especially tokens that have hybrid functions, or represent completely new phenomena. Entrepreneurs in the latter case will always be confronted with uncertainties of how the regulator might retroactively classify the token. To provide regulatory certainty to entrepreneurs, some jurisdictions have started to offer governmental sandboxes to guarantee innovation while allowing for a process of regulatory learning.

Purpose Perspective:
As opposed to tokens that represent existing assets or access rights to an asset or services someone else owns, tokens can also be programmed to incentivize a new way of collective value creation. They can be used to incentivize individual behaviour or contributions to a collective goal of a group of people, if and when you bring proof of contributing to a collective goal. Bitcoin and other protocol tokens are a good example for such purpose-driven tokens. As such, purpose-driven tokens are similar to rewards programs or loyalty programs, and can be used for nudging purposes. While nudging and rewards programs are nothing new, the advent of cryptographic tokens has spurred a lot of innovation around purpose-driven tokens that incentivize behaviours that are benfficial to the general public, like CO2 tokens, or time bank tokens, etc. Such tokens could prove to be a new incentive tool, to reward individual contributions on the fly, as opposed to the current after-the-fact disincentives most institutions use today (read more on purpose-driven tokens here).

Token Supply Perspective:
Poker chips have a fixed price and unlimited supply. A casino can always produce more chips. Many of the tokens listed on coinmarketcap, especially those that were used for early token sales to raise funds, have limited token supply, possibly because no investor wants to buy into an unlimited supply to protect their investment. This is especially true for tokens that represent currencies, assets, or equity, or debtissuance of tokens. For tokens that represent an access right, the number of tokens is usually limited to the maximum capacities and frequency of the access provider. The only limit is the capacity of a system, such as the capacity of the busses of a public transport network, which can always be extended if necessary, most often including a time lag. What makes at money or a precious commodity like gold valuable is that it is limited in supply. Any tokens with a limited supply could therefore be de-facto stores of value, depending on the expected durability of value, as well as the short-term volatility of their price (read more: Part 3 – Stable Tokens).

Token Flow Perspective:
Another dimension revolves around the question of the token flow. Tokens might be created for a single purpose and destroyed when used. In this case, they flow in a straight line from source to sink. Examples thereof are casino chips that can be used within the realm of the casino and are issued against at currency. Once a player leaves the casino, they can cash the chips back to local at currency. Any transportation ticket that pays for access to a system and then expires after one time of use, or a er a period of time, is another example. Their price is determined by the service provider (low volatility) and the supply is not limited, or limited to the infrastructural conditions. Such tokens must be destroyed by consumption or expiry date, to complete the cycle. On the other hand, tokens that can be exchanged back and forth inde nitely, without an arti cial expiration condition, can be said to have a circular ow. When you transfer the token (currency or piece of art), someone else receives the token. This person can spend the token to buy something or trade the token on an online exchange for another token. Tokens with a circular token flow will only sink when you lose your private keys, or if the physical underlying asset is accidentally destroyed.

Temporal Perspective:
Another question when designing a token is whether the token has an expiration date. Any fungible token might be programmed in a way that it expires after a certain date to prevent hoarding of the tokens. Practically speaking, the token would expire. Technically speaking, the token would change state. Bonus points of loyalty programs usually come with an expiry date. In the past, some regional currencies, like the Wörgl Schwundgeld (Austria) in the 1930s, experimented with an inbuilt deflation of their currency to prevent hoarding and inflation. The currency was introduced as a parallel currency, that could only be spent in the region of Wörgl. By losing one percent of its value each month, individual spending was encouraged while saving was disincentivized. This measure was introduced to combat a country-wide deflationary policy and helped with both unemployment numbers and infrastructure investments.

Full text and high-resolution graphics available as paperback & ebook: Token Economy, by Shermin Voshmgir, 2020