Disclaimer: This article should not be taken as an investment advice to buy or sell any securities or cryptocurrencies.
Investors and financial analysts are constantly on the lookout for a simple yet informative indicator to assess stock and commodity prices. In recent years, cryptocurrencies or tokens (the two terms are interchangeable in this article) have become the latest focus of this search. An optimal indicator should reflect the position of the market valuation of the asset relative to its fundamentals. Is it overvalued or undervalued? When is a good time to buy or to sell?
This article reviews the price-to-earnings ratio (a widely used indicator in stock market) and the network-value-to-transactions ratio (a metric recently developed for the crypto market). I evaluate these two ratios in the context of the cryptocurrency market, discuss their limitations, and, ultimately, propose an alternative indicator: the price-to-utility (P/U) ratio. The latter is, in my humble opinion, better suited for the valuation of cryptocurrency.
The figure below shows the P/U ratio of Bitcoin. The P/U ratio clearly identifies the two big Bitcoin bubble periods at the end of 2013 and 2017, where the P/U ratios were way out of its normal range between 50 and 200.
Token Valuation Indicators: Overview
Unlike the market valuation of an asset, which can be precisely measured by its market cap, the fundamentals of the asset can be difficult to determine.
The price-to-earnings ratio adopts the annual earnings of the company (or equivalently, the dividends allocated to stockholders) as the fundamentals of the company’s stock. It is a simplistic but effective proxy.
Most tokens in the crypto market, however, do not bear dividends. As a result, a new approach was proposed to gauge the fundamentals of tokens by the value of transaction volume over the past 24 hours on the blockchain. This approach became known as the network-value-to-transaction ratio. Here the fundamental value of a token derives from how frequently it is used for transactions, i.e. from its function as a medium of exchange.
Such a ratio overlooks the token’s function as a store of value. The store of value is one of the three functions of the major sovereign currencies (the other two are medium of exchange and unit of account).
In this article, “store of value” refers to the function of a token to keep or increase its purchasing power over time. It should be noted that due to the high volatility of the token price, most cryptocurrencies, except for stable coins, do not serve as a unit of account.
To summarize, a comprehensive assessment of the fundamentals of cryptocurrencies should incorporate their functions as exchange medium as well as value storage. Both are captured by the Price-to-Utility (P/U) ratio.
The detailed description of the P/U ratio is shown below. For those familiar with the price-to-earnings ratio and the network-value-to-transaction ratio, please skip the next two sections and continue directly with the section on Price-to-Utility Ratio.
The price-to-earnings ratio, henceforth the P/E ratio, compares the stock price to its earnings per share. It is calculated by dividing the current stock price by its annual earnings per share (EPS). For example, the stock price of Amazon on Dec 31 2018 was roughly $1500 and its EPS for that year was around $20, yielding a P/E ratio of 75.
The P/E ratio indicates how many dollars investors are willing to pay today in exchange for one dollar annual EPS in the future. In the example of Amazon, investors were willing to pay $75 for $1 annual EPS.
P/E Ratio = Stock Price/EPS
= Stock Price/(Annual Earnings/Outstanding Shares)
= Market Cap/Annual Earnings
The P/E ratio measures the company’s market cap relative to its annual earnings and sheds light on the valuation of the company’s shares. Generally, analysts compare P/E ratios of various stocks and recommend to buy when the ratio is low and to sell when the ratio high.
Although the P/E ratio is an established tool that has long been providing investors with insights into financial markets, it has limitations.
The EPS is calculated using the company’s earnings of the past 12 months (aka trailing EPS). The historical revenue, however, does not reflect the future. That is to say, this metric does not factor in the growth of the company. A high P/E ratio (i.e. a high stock price relative to its earnings) does not necessarily imply the stock is overvalued. It could also be due to the market’s anticipation of the company’s rapid upcoming growth.
Certain analysts rely on reported revenue forecasts (aka forward EPS), but this approach involves uncertainty and is susceptible to report manipulations.
Nonetheless, the P/E ratio provides a simplistic yet practical measure for a quick evaluation of the stock price.
Overall, the P/E model does not work in the realm of cryptocurrency. Unlike companies, most blockchain projects do not have earnings. Although it should be noted, that airdrops (e.g. Ether token holders in the ICO craze) and forked tokens (e.g. Bitcoin Cash, Bitcoin Gold and Bitcoin Diamond etc for Bitcoin holders in the hard fork hysteria) can be seen as dividends to token holders. But this type of earnings is neither predictable nor sustainable.
Researchers suggested transaction fees as a proxy for the earnings. Using transaction fees as a substitute is misleading, because these fees are earned by miners/blockmakers and not by token holders.
Furthermore, using transaction fees could lead to ludicrous results. To illustrate: higher transaction fees lead to a lower P/E ratio and, consequently, to an undervaluation of the token. In other words, Bitcoin becomes undervalued (with decreasing P/E ratio) when transaction fees skyrocket. It is no secret that high transaction fees impede a more universal adoption of Bitcoin.
The figure below on the left demonstrates the P/E ratio of Bitcoin over the past five years, using transaction fees as the proxy of Bitcoin earnings. If newly minted Bitcoins (block rewards) are taken into account, the P/E ratio is in the range of 5–125 (see the figure below on the right). Again, it is important to bear in mind that the P/E ratio is meaningless to Bitcoin buyers and sellers, since transaction fees and block rewards are not allocated to them.
Blockchain projects are unlike public companies that allocate net profit to their stakeholders. Consequently, traditional P/E ratio does not apply to token valuation.
There are some exceptions. Among them, the Binance Coin (i.e. BNB) is a notable example. Binance is a cryptocurrency exchange platform. It uses 20% of its quarterly profit to repurchase its native tokens and burn them. The repurchase and burn program is equivalent to the allocation of dividends to BNB token holders.
The table below shows the P/E ratios of the BNB token on the dates of the burn event.
The total BNB supply (Total Token Supply or TTS) was 200 million and 136 million (Token in Circulation or TiC) was released when the project was launched in the summer of 2017. 64 million locked BNB were to be released in the following four years (16 million BNB released per year).
Note that there is a significant rise of P/E ratios when calculating both measures (using TiC and TTS) in the first quarter of 2019. The climb indicates an overvaluation of the BNB token compared to its historical record. Yet, as mentioned above, the P/E ratio does not factor in the expectation of growth. In fact, the rise of the P/E ratio can be attributed to the reignition of the Initial Exchange Offering (IEO) and Launchpad on the Binance platform, as well as the launch of the Binance Blockchain and Decentralized Exchange (DEX) in the first half of 2019.
P/E ratio may have served the stock market well, but in the case of the crypto market it either fails to apply entirely or leads to false implications due to the fundamental differences between public companies and blockchain projects.
Metcalfe’s Law and Network-Value-to-Transactions Ratio
Various researchers employ Metcalfe’s Law to study the fundamental value of Bitcoin. Metcalfe’s Law stipulates that the value of a network is proportional to the square of the number of its users (or, in this case, token holders). The number of Bitcoin users can be proxied by the number of Bitcoin unique addresses. Metcalfe’s Law captures long-term value of a token by measuring the number of its users. Nevertheless, research has established that, in the short run, the validity of Metcalfe’s Law for cryptocurrencies is questionable.
For improvement, I propose to also take into account the transaction volume per unique address. Such a modification would make Metcalfe’s Law-based models more suitable for token valuation analyses. This can be explained by the fact that the number of users reflects the long-run value of the token, while the transaction volume per unique address mirrors the daily usage of the token. Consequently, the transaction volume per unique address should be factored in to account for the short-run price variation. Empirical economists are welcome to look further into this issue.
Network-Value-to-Transactions (NVT) ratio, introduced by Willy Woo, may serve as an analog of the P/E ratio for the crypto market. The NVT ratio is defined as the ratio between network value (i.e. the market cap of the token) and transaction volume in USD over the past 24 hours. A high NVT ratio indicates that the market cap of the token outpaces the transacted value on its blockchain, indicating speculation. A low NVT ratio caused by an increasing transaction volume indicates a high usage of the token, which, in turn, serves as a signal to buy the undervalued token. Note that with the development of the scaling solution (e.g. Lightning network for Bitcoin and Raiden network for Ethereum) the off-chain transactions should also be taken into account.
NVT Ratio = Network Value/$ Transaction Volume
= Token Price*Token Supply/Token Price*Transacted Token
= Token Supply/Transacted Token
= 1/Token Velocity
The limitation of the NVT ratio lies in the assumption that the fundamental value of cryptocurrency derives only from its function as a medium of exchange that is represented by token velocity. The usage of cryptocurrencies as a store of value is completely overlooked in the model.
To illustrate the point: an increase of the NVT ratio due to a reduction of the transaction volume does not necessarily imply an overvaluation of the token. It could also be the result of an increasing number of long-term hodlers hoarding more tokens, thus causing a decline in transaction volume. In other words, the NVT ratio offers a glimpse of the picture and is not particularly informative regarding buying and selling decisions.
Willy Woo himself admits that the NVT ratio is not an effective tool to detect a price bubble ex ante. Instead, it is better suited to discern between price collapse and price consolidation retrospectively. Still, it failed to explain the 2017 crypto bull market when the NVT ratio was below its normal range, thus indicating undervaluation. The NVT ratio also failed to identify the 2018 crypto bear market when there was clearly a market collapse while the NVT ratio remained in its normal range. Despite the shortcomings of the NVT ratio, Woo earned his reputation as a pioneer of token valuation related research and deserves a big shout-out.
Token utility is defined as
token utility=(token velocity*staking ratio)/dilution rate,
where the dilution rate measures the annual growth rate of the token supply. Assuming no other differences, a high dilution rate makes the token less appealing, thus leading to a lower token utility. The figure below demonstrates that the dilution rate of Bitcoin has significantly decreased over the past decade. As of May 2019, the dilution rate is around 3.8 percent. A halving event of the BTC supply is expected in one year. By that time, the dilution rate will have fallen below 2%, which is the common inflation rate targeted by major central banks.
Token velocity measures the percentage of tokens transacted over the past 24 hours relative to the token supply. A large token velocity signifies an increasing usage of and demand for the token. It leads to a high token utility.
As for Bitcoin and some other major cryptocurrencies, quite a substantial number of tokens are inactive for more than a year. These inactive tokens serve as a store of value and can be referred to as “staked tokens”. Active tokens, in turn, can be called “liquid tokens”. The staking ratio is defined as the ratio of staked tokens with respect to the total token supply. A high staking ratio suggests that more users are placing long-term faith in the blockchain project. The figure below shows the BTC staking ratio under different staking periods. The trend going forward appears to be that more bitcoins are staked, for longer, as time progresses.
Token velocity represents the utility of the token in its role as a medium of exchange that is demanded by daily active users (DAU). The staking ratio represents the utility of the token as a store of value, demanded by long-term hodlers (LTH). In the case of Proof-of-Stake (PoS) blockchain projects a fraction of tokens is staked in the system. For Proof-of-Work (PoW) blockchain projects, “staked tokens” are tokens that have been kept in inactive accounts for more than one year.
In short, a low dilution rate combined with a high token velocity and a high staking ratio results in a high token utility. The figure below illustrates the token utility of Bitcoin using BTC 1-, 2-, 3-, and 5-year staking ratios.
To accommodate the different natures of blockchain projects, we could add a weight α ∈ [0,1] in the formula.
token utility=(token velocity*staking ratio^α)/dilution rate.
For general-purpose PoS projects: α=1. For projects focused on payment (e.g. OmiseGo, Stellar, Zcash): α should be smaller than 1 to place more weight on token velocity. Note that the NVT ratio is a special case where the token utility is reversed with a normalized dilution rate and an overlooked staking ratio (α=0).
To smooth token utility, one could, for example, replace the token velocity over the past 24 hours with the average token velocity over the past 90 days. It is difficult, if not impossible, to develop a simultaneously simple yet accurate token valuation methodology. Thus, for now, I will strive to keep the formula as elegant and simple as possible.
As the Figure above illustrates, the token utility of Bitcoin has been increasing dramatically over the past five years. When comparing token utility with the 5-year staking ratio (blue line in the figure below) to the BTC price (red line in the same figure), it appears clearly that token utility is a leading indicator of the BTC price. An increase in token utility that results from the rise of the token velocity and of the staking ratio, along with the decrease of the dilution rate implies that fewer liquid tokens circulate in the system than are in demand. Increasing demand boosts the BTC price until a new supply-demand equilibrium is achieved. This explains the price hysteresis.
The recent crypto spring in the first half of 2019, along with a doubled crypto market cap, is due to the increase of the BTC 1-year staking ratio. The hike of the BTC 1-year staking ratio will ultimately result in the hikes of 2-, 3-, and 5-year staking ratios over the next few years. A reduction of liquid BTC tokens together with an increasing BTC velocity led to the end of the bearish market of 2018.
To give an example of PoS project: assuming a blockchain project issued a total of 100 million tokens with a dilution rate of 5%. 40 million tokens are staked in the system, rendering 60 million liquid token in circulation. The transaction volume over the past 24h is 15 million token. Based on these numbers, the following calculation can be done:
token velocity=15 million/100 million = 15%
staking ratio=40 million/100 million = 40%
token utility=(token velocity*staking ratio)/dilution rate=1.2
For PoS projects, increasing staking rewards by setting a high dilution rate has two counteracting effects. On the one hand, a high dilution rate lowers token utility. On the other hand, high staking rewards incentivize LTH to stake more tokens, which leads to a high staking ratio and ultimately results in a high token utility.
The table below compares token utility of blockchain projects with different designs. Project B has a higher dilution rate than Project A, yielding lower token utility. Project C has lower token velocity than Project B, implying a lower DAU of the blockchain. Project D has a lower staking ratio, implying a lower ratio of LTH.
A high dilution rate combined with a low token velocity and a low staking ratio leads to low token utility.
With token utility defined, P/U ratios of different blockchain projects can be easily obtained. The figure below demonstrates a zoom-in version of BTC P/U ratio. As can be seen, the Bitcoin P/U ratio lies between 50–200 in normal times, above 200 in bubble ages and below 50 in crypto winters (as illustrated by the horizontal lines in the graph).
Comparing P/E ratios of stocks in different sectors is like comparing apples to oranges. Bear in mind that the same goes for the P/U ratio in the crypto market. Blockchain projects address different issues, provide diverse services and target various groups of users. For example, Bitcoin has two major functions: it serves as a decentralized peer-to-peer payment system as well as a storage vehicle. As a comparison, the value of Ethereum is reflected mostly by the daily active usage of its smart contract platform.
Caveat: There is no single metric that can provide a comprehensive overview of the token valuation. Price-to-Utility (P/U) ratio is not designed to spot signs of market booms and crashes but it can be instrumental when comparing the valuations of tokens with similar blockchain features.
The author of this article, Yulin Liu (firstname.lastname@example.org), currently leads research on token economics and governance system at DFINITY. He also serves as Affiliated Economics Professor at Huazhong University of Science and Technology.
Yulin specialises in monetary theory, bank supervision, cryptocurrency, token economics and blockchain governance system.
He holds a Master of Science in Quantum Computation and a Ph.D. in Economics from ETH Zurich. Yulin was a visiting scholar at the European System of Central Banks and has been invited for talks at major central banks and conferences worldwide.