Even the most ardent cryptocurrency enthusiast will usually admit, in a vulnerable moment or after a few drinks, that parts of the cryptocurrency market are likely overvalued, and perhaps enormously so. But the question of precisely how cryptocurrencies should be valued is rarely discussed at HBS. Given the level of investment in the space, this question clearly needs to be answered. The following article will attempt to articulate a logical approach to valuation that does not simply depend on the “Greater Fool Theory” that somebody, somewhere will be willing to buy these digital assets at a higher price without considering the fundamentals.
When you buy into most cryptocurrencies, you are generally buying a unit of economic value that can later be redeemed for a product or service. For example, Ripple XRP’s token is planned for use in fast cross-border payments, Ethereum’s Ether for powering other cryptocurrencies and computing power rental, Monero for private payments, and Filecoin for peer-to-peer file storage rental. In each of these cases, the cryptocurrency token has value because it will at some point be used by its eventual owner to perform one of the above functions. The current holders are generally speculators who believe that eventually, someone will actually want to use the token for its “true” function, and at that point the speculator will sell it to the person who needs to use it. Thus, after stripping away the speculation, we should be able to measure the “use value” of each cryptocurrency token in performing these functions, rather than simply their “exchange value”.
So where do we start? A discounted cash flow (DCF) model, typically used to value equity investments, will not work. This is because cryptocurrency tokens will not themselves generate recurring profits or free cash flows; rather these tokens are transactional units within a protocol that only produce cash flows when used, redeemed or sold. Cryptocurrency tokens also cannot be valued using multiples of comparable assets because they behave unlike existing traditional or alternative asset classes in existence.
The way to value cryptocurrencies, as Vitalik Buterin, founder of Ethereum, and Chris Burniske of Placeholder VC have argued, is to use macroeconomic principles. To do this we can start with the “basic equation” of monetary economics:
MV = PT
- M = Money supply
- V = Velocity of money in the economy (defined below)
- P = Price level of goods and services
- T = Transactions or total GDP
MV = PT posits that the amount of money in the economy “M” (i.e. in the U.S., the total number of dollars worldwide), multiplied by the velocity of money “V” (defined in detail below) will be equal to all the transactions in the economy “T” (often proxied as GDP), multiplied by the price level of goods in the economy “P” (this factor scales the transactions up if prices are high, and down if prices are low).
For cryptocurrency tokens this equation can be rearranged and substituted as follows:
SV = PT
Therefore 1/P = F = T / (SV)
- F = “Fair value” price of cryptocurrency token (equivalent to 1/P above)
- T = Total value of token transactions per year
- S = Total supply of circulating tokens (equivalent to M above)
- V = Velocity of token supply per year
In the cryptocurrency rearrangement above, the terms deserve some extra unpacking. “T” and “S” are intuitive. The transaction value “T” represents the total annual transactions using the token in a given year. For example, “T” for Monero — the token in a private transaction protocol, increasingly used by criminal networks and reportedly even the North Korean government — might be the total market size of illicit value transfers, multiplied by Monero’s expected market share of these transactions. The equation suggests that if transaction value “T” is high, then the value of each transacting token will need to be correspondingly high given the fixed volume of tokens. Supply “S” represents the number of tokens issued by the protocol. Moving to the better-known example of Bitcoin, there are currently over 16 million Bitcoin issued, with a theoretical maximum of 21 million. If the supply of tokens “S” is low, then price per token will also need to be high to reach the transaction volume. Clearly, therefore, a high “T” and a low “S” will result in a high-value token.
The part of the valuation framework that is less often considered — velocity “V” — merits more detailed consideration. The velocity of cryptocurrencies measures the number of times the same crypto token is used within a given time period. This principle is borrowed from monetary economics, where “V” measures the number of times the “same dollar” is transacted in a time period. Velocity has an inverse relationship with currency appreciation. For example, if a banking system becomes more efficient, meaning money is transacted more quickly, the velocity of money increases, because each dollar can be used more times per year. All else being equal, this is inflationary, since more money is available to buy the same amount of goods and services (i.e. prices of goods go up and the value of each unit of money decreases). In cryptocurrencies, there is a similar effect.
High velocity of cryptocurrencies — driven among other things by fast transaction times, highly liquid markets, or low user working capital needs — is inflationary, because it means the same number of tokens are recycled more often within a given time period and therefore effective supply is increased. Conversely, if velocity were to decrease, the value of the each token should increase.
Consider a hypothetical cryptocurrency with 1,000 tokens that is used to transact $1,000 per year. If each token were only used once per year (and therefore its velocity equals 1), the market value of each token would be $1. If each token is instead used twice per year (and therefore its velocity equals 2) while the total value of transactions remains $1,000 then each token needs to be worth only $0.50. Hence, high velocity reduces price per token.
This leads to a counter-intuitive conclusion: the longer and more inefficient a cryptocurrency’s transactions are, all else being equal, the higher its value will be, and vice versa.
As we will see below, this possibility could be critical to the valuation of some cryptoassets, such as Ripple’s XRP token, which has transaction times of less than ten seconds. I argue that this efficiency of exchange, although it is the primary selling point of these assets, pushes down their value, because their high velocity increases effective supply at any given time.
Putting the above together, and building on the work of Chris Burneske, I have produced a simple model to estimate the fair value of any active cryptocurrency: https://goo.gl/H1E7fp. Instructions are listed in the second tab. The inputs can be changed to value any cryptocurrency, since fundamentally the model simply calculates the total market “T”, and divides it by the token supply “S” and the velocity “V”.
After playing with the assumptions, it becomes clear that the value of a cryptocurrency is highly sensitive to the velocity of its tokens. That is, given a certain market size, penetration and token supply, the better the infrastructure gets, and the more liquid markets become, the lower the token value will fall.
For example, consider Ripple’s XRP token, which aims to facilitate interbank transfers. Let’s assume all speculating self-styled “hodlers” (crypto-talk for “holders”) fell away and XRP captured 100% of the market in interbank payments, worth $125 quadrillion annually. If transaction times remained at 6 seconds, and market liquidity generated time between uses of 12 seconds, the fundamental value of XRP would only be $0.54, one quarter of its year-end 2017 price. If, on the other hand, Ripple could capture the entire interbank market and simultaneously slowed transaction times from 6 seconds to two minutes, and confined market liquidity to a post-transfer token exchange of five minutes, the fair value of XRP would increase dramatically to $21.62. Using the same reasoning for any cryptocurrency, large or small, demonstrates the same point: velocity is crucial. Bitcoin is equally sensitive to the effects of velocity, but the possible range in its fundamental value are even greater, since there is such a large proportion of “hodlers” who view it as a store of value rather than a transacting token.
Of course, right now the over-exuberance and immaturity of the market means that we may be orders of magnitude away from “fundamentals” or “fair value” for plenty of cryptocurrencies. However, once the market stabilizes and equilibrium is reached, a value investing approach should be rewarded. This approach is as follows. First, aim to maximize transactions “T” by investing in tokens that target large markets, with a good value proposition, an experienced team and limited competition. Second, minimize supply “S” by seeking out those with small or limited token supplies. Finally, and crucially, neglect at your peril the importance of minimizing velocity “V”, even if it means investing in illiquid markets and technologies that are likely to be less efficient.
Additional disclosure: I have invested in several of the cryptocurrencies and protocols mentioned above. This article does not constitute investment advice.
Sasha Kaletsky (HBS ‘19) graduated from the University of Oxford in 2014. Prior to HBS, he launched and managed various UK cities for Uber. Follow him on Twitter at @SKaletsky.