A number of laws and regulations, such as antitrust laws and financial regulations, are informed by the relative economic size of the companies or assets under investigation. For the blockchain, this means that cryptoassets – i.e. digital instruments of economic value that are developed and traded on blockchain networks (e.g. cryptocurrencies, tokenised securities, crypto-derivatives, etc.) – with larger market shares are likelier targets for law enforcement or regulation.
Properly measuring the economic footprint of cryptoassets becomes imperative. However, measuring the relative economic size of cryptoassets has proven challenging for multiple reasons.
This paper. available here, presents the first systematic examination of the economic footprint of cryptoassets and their constituent actors, with the goal of providing comprehensive guidance into the size of the crypto-economy.
The article proceeds in three steps:
Part II explains the function and challenges of market share calculation.
While some rules and obligations apply uniformly across industries, many are contingent on the relative size of the regulated subjects, meaning that smaller entities may incur different regulatory treatment compared to larger entities. Antitrust law is the prime example, with many rules being triggered only after market power is established, whose main (but not exclusive) determinant is the relative size of the challenged firm in a defined market.
A common way of determining market size is to look at market shares. Market share metrics typically include sales—whether by revenue, units or other indexes – and various measures of production capacity. However, different economic sectors may require the use of different metrics. In the financial sector, which encompasses cryptoassets, the relevant metrics have included: total assets, deposits, and loans in the retail banking market; transaction value and volume in the stock exchange market; number of cards and volume of card transactions in the credit/debit card market; and “league table” rankings in the investment banking market.
Even when relevant metrics have been identified, it is not at all clear which one should be decisive. This selection can be influenced by the nature and purpose of the market share assessment. In merger cases, where the assessment is forward looking, courts and authorities have tended to rely on metrics that reveal firms’ future economic strength; whereas with anticompetitive agreements or in monopolisation cases, the emphasis is on metrics that reveal firms’ market shares at the time of the anticompetitive conduct.
Given the heterogeneity of available metrics, no single metric should be presumed to be superior to the others. In fact, novel markets may be characterised by unique metrics. The only meaningful approach for cryptoassets is to list the relevant metrics in cryptoassets’ markets, and analyse what they mean and what they show about the economic strength of cryptoasset market actors. The rest of the paper is devoted to developing this approach.
Part III looks into the calculation of inter-asset market shares.
One metric that may appropriate for cryptoassets is inter-asset calculation – i.e. the measurement of market shares of cryptoassets relative to other cryptoassets. The most popular inter-asset comparative method is market capitalisation, which reflects cryptoassets’ economic footprint as a function of their market price multiplied by the circulating supply. Informative as it may be at first glance, market capitalisation is problematic because it treats all cryptoassets as homogeneous and glosses over numerous influential considerations regarding their size, use and health.
To address these weaknesses, the authors disaggregate different cryptoassets and cluster them into like-kind groups, and then propose a number of alternative metrics of cryptoasset economic strength.
The cryptoasset market is highly diverse, and includes tokens intended as straightforward currencies, (pseudo) or actual tokenised equity, scarce representations of some commodity (e.g. storage and computing power), access keys for a given application or service, and representations of some external asset like gold. Since these are all diverse functions, it is worth devising a taxonomy of cryptoassets. This can be done in a number of ways, all of which have limitations.
- A first approach focuses on the function of a cryptoasset. This can distinguish, for example, between cryptocurrencies, crypto-commodities (e.g. access rights to some digitally scarce commodity like storage or computation) and tokens (access rights to some finished digital good or service, like a social media platform or a prediction market).
- A second approach focuses on where the information necessary for the cryptoasset to perform its function resides. Such an approach can distinguish between cryptoassets that rely on endogenous references, which derive entirely from internal data, and cryptoassets based on exogenous entries, which refer to external data which must be validated by third parties.
Regardless of the specific taxonomy employed, the significant heterogeneity in both form and function among cryptoassets is cause for disaggregation. Meaningful comparisons should be made within peer groups, rather than across a universal sample of cryptoassets.
Once one has identified the relevant group of comparable cryptoassets, the measurement of cryptoasset economic strength will typically rely mainly on market capitalisation. However, and in addition to the limitations discussed above, this method is often undermined by the unavailability and unreliability of data. A particular challenge in this respect concerns the measurement of available supply. This challenge may arise as a result of supply arbitrariness – i.e. the total supply of a cryptoasset may be known in advance, but the time of release remains unknown and often at the discretion of a single entity –, inert supply – i.e. assets that are in circulation are highly illiquid, which may lead one to overestimate an asset’s truly active and liquid supply –, and supply inheritance – i.e. the difficulty of measuring the overall supply contained in the forks of a previously unified blockchain.
Alternative measurements of cryptoasset market shares, which all pose their own difficulties, include transactional capacity, transaction value, and the aggregate value of transaction fees spent to transact on-chain in a given timeframe.
Part IV then tries to determine how to calculate intra-asset market shares.
Intra-asset calculation refers to the measurement of market shares of the actors that make up cryptoasset networks, namely actors or mechanisms that enable cryptoassets and underpin their operation. This class includes numerous actors, but the authors focus on the ones that have attracted the most scrutiny and interest: miners (including mining pools) and crypto-exchanges (including wallets).
Mining technically refers to the process of creating new denominations of cryptoassets. Miners are essential in cryptoassets whose consensus mechanism relies on commitment of resources—whether processing power, storage space, time or other—and where the reward for committing the necessary resources comes in the form of a denomination of a cryptoasset, e.g. Bitcoin. In effect, mining has become synonymous to adding (“finding”) new blocks on a blockchain, because in the majority of the most popular cryptocurrencies a denomination of the cryptocurrency is created as reward whenever a new block is added to the respective blockchain.
There are a number of metrics to measure the relative market relevance of miners. Output metrics calculate market shares based on an appropriate standardised unit of production (e.g. blockcount, the number of blocks a miner adds to the blockchain). Capacity metrics look at the rate at which resources can be committed to production output (e.g. hashrates, the cryptographic value that miners have to compute in order to find a new block and add it to the blockchain).
Market share calculation by revenue is also common, since revenue shows the portion of the market captured by unit of currency spent by consumers. Revenues in blockchain can come from block rewards or transaction fees. Another metric, related to the fact that revenue may not be under the control of miners, is profitability. Profitability may be particularly relevant when the blockchain’s internal rules can be shaped to increase the profitability of those with greater mining power (e.g. pools) at the expense of profitability of the individual miners. Because solo mining is often inefficient, it is common for miners to aggregate their resources to create mining pools; one possible metric of mining pools’ market share is miner count, i.e. the number of miners that commit their resources to one or more mining pools.
Crypto-exchanges are service providers that allow the trading of cryptoassets—whether with other cryptoassets or with fiat currencies. Crypto-exchanges commonly perform wallet functions, which makes them the custodians of users’ assets, and allows users to manage their addresses and to transfer funds between them.
Exchanges are characterised by a number of metrics. Trading volume looks at the total value of the assets traded in an exchange at a given time. This metric is relevant for market share calculations because the role of the exchange market is to enable the trading of assets, and trading volume shows what part of the total traded (exchanged) asset value is taking place on a given exchange. Another common metric is the number of transactions performed in a platform at a given time. In the context of crypto-exchanges, this metric represents the total number of trades between cryptocurrencies, and between cryptocurrencies and fiat currencies. However, care must be taken regarding both these metrics – reported data can be misleading, and listed transactions manipulated. A last metric measures the number of trading participants, i.e. the number of individuals who trade on an exchange through direct access to the trading platform. While of dubious relevance in financial markets where users rely on brokers, crypto-exchanges interact with users, so measuring user numbers may provide am accurate picture of the market.
This is a bold, very technical attempt to map out the various ways to measure the relative economic power of various cryptoassets.
While recognising that the goal of the paper is limited to identifying methods for measuring market power, and that the paper also seeks to provide guidance beyond antitrust, I nonetheless think that it would have been useful if the paper had included a discussion of how to define cryptoasset markets. The reason I say this is because the paper focuses mainly on the difficulty of using different metrics and on the practical reasons why one might choose to rely on a specific metric instead of another. This approach presupposes that we will know what the relevant market is, and that the main difficulty will be measuring measures. On the basis of this and the other papers I have reviewed, I think this is a rather optimistic assumption.
Please note that I am not saying that the authors make any such an assumption; on the contrary, I have no doubt that they are aware of the difficulties inherent in defining cryptoasset markets, and I think they try to briefly address these difficulties when talking about how to distinguish between various types of cryptoassets.
My point is merely that an analysis of how to select the best methodology to measure market power that relies solely on considerations related to the practical feasibility and difficulty of using such methodologies will necessarily be incomplete, because the selection of a methodology will necessarily depend on what the relevant market is and how we are able to define it. Furthermore, a market definition approach that is based on disaggregating the various components of cryptoassets is reminiscent of defining a market by reference to product characteristics, and we all know that this is not an optimal way to define a market, to say the least. As such, this is a matter that I would like to see addressed in future work.