Frameworks for assigning real value to crypto-assets (Part I)
Ask anyone and they’ll probably say:
- Look at the website
- Look at the team
- Read the whitepaper
- Think they can actually do the things in the whitepaper?
This has served okay, mostly because we are all massive speculators, and speculation is what has driven these prices to their sky high valuations.
Moving forward, assigning real value to currencies will require a more rigorous approach. In this article I present a couple of frameworks our fund uses to value crypto. I’ve split them up into two different categories.
Absolute Valuation: where value will be determined based on an asset’s fundamentals
Relative Valuation: where value will be determined based on an asset to asset comparison
Part I will deal with absolute valuations
Lets start with the Absolute Valuation:
1. Discounted Cash Flow Analysis (DCF)
Since we have no way of reliably estimating future cash flows or the discount rate to use given the large variance in risk level, we change the inputs by replacing projections for sales, D&A, etc. with inputs in John Maynard Keynes’ classic equation of exchange to derive each year’s current utility value. Chris Burniske has written on the topic and proposed that we can then discount this future utility back just like we would in a DCF to find what the utility would reasonably be worth now. However, we will not deeply explore the actual implementation of the model since crypto markets are too irrational to base investment decisions on these factors. Rather, we will use the equation of exchange as a currency model to highlight important facets of value behind a token.
Equation of Exchange:
This equation was initially proposed by Cambridge University Economists Economists Alfred Marshall, A.C. Pigou, and John Maynard Keynes. It has been adapted for crypto markets below.
M = total number of coins in circulation
V = velocity of the coin (average number of times it changes hands) in a day/year
P = inverse of the token’s price, since we are looking at price of the network in terms of a token
T = economic value of transactions in a day/year
This should make sense conceptually: if there are M coins and they switch hands V times, the transactional volume is M*V. If this results in T dollars of economic value, the dollar value price can be represented by T/(M*V). This means that the inverse of the price, P, is (M*V)/T.
Some other relationships can be defined from these terms:
H represents the time a user holds a coin before using it to make a transaction and C is the cost of buying the currency. A new equation then arises:
The left side of the equation, total number of coins times price, is the market capitalization. The right side represents the economic value transacted per day multiplied by the time the user holds a coins before using it to transact. Rearranging for the price of the coin, C, we can see that the price of a coin will increase in one of three situations:
- Economic value transacted per day increases
- Average time a coin is being held increases
- Number of coins in circulation decreases
The second point can be confusing and there are some differing opinions on the subject. On one side, it is believed that for utility tokens their value becomes higher the less tokens are in circulation (i.e. aggregate demand is higher and people hold longer due to the store of value use case). On the other side, there are some that would more highly value a coin based on its means of exchange. In this case, higher velocity would mean higher value. Therefore, we suggest doing an analysis of each coin to see what group its technology and investors fall in and tailor the model. Regardless, velocity is a very important metric to be looking at when analyzing a digital assets value. A great in depth explanation can be found here.
Using the equation of exchange in combination with something called a S-shaped market adoption curve is the most common application of this in practice. The full picture can be created by integrating a qualitative fundamental analysis as well. We won’t go into detail here since other resources do it justice better, and basing valuations on a variation of DCF in a volatile market are risky, but models can be found here credit to reddit user arsonbunny:
- Part 1 — A fundamental quantitative valuation of Bitcoin
- Part 2 — A fundamental quantitative valuation of REQ (Request Network)
- Part 3 — A fundamental quantitative valuation of VEN (VeChain)
- Part 4 — A fundamental quantitative valuation of FUN (Funfair)
- Part 5 — A fundamental quantitative valuation of ETH (Ethereum)
2. Dividend Discount Model (DDM)
For digital assets that offer dividends, such as exchange coins (in the form of a fee share), a DDM can be used. The growth rate, g, would be the expected growth rate of exchange trading volume multiplied by the fee share percentage. There will obviously be a wide projection of exchange growth rates, but this model can provide some insight as to which assets are vastly mispriced. The required rate of return, r, is equal to returns in the overall crypto index.
Instead of issuing a dividends through a fee share due to legal concerns, some projects might burn tokens — similar to a share buyback. A token burn, like a buyback, is done to reduce the total supply of a tokens available on the open market, increasing the value of the asset for investors. Some tokens might even do both.
Keep an eye out for our analysis using this method for some of the most popular exchange coins out there: Binance’s Coin (BNB), Kucoin Shares (KCS), COSS (COSS), and Cryptopia Fee Shares (CEFS). For a sneak peak, to solve the issue of varying growth rates (especially when g > r in the beginning), we will be using the following formula to value them:
3. Cost of Production
This is a completely new model that has not found its way from traditional finance. It is important to understand that part of a digital asset’s value comes not only from buying them or exchanging them, but also from producing them through processes like mining. Adam Hayes investigates this concept in his paper modeling the price of bitcoin. Essentially, he found that three variables determine 84% of the value of digital assets: computational power, rate of coin production, and relative difficulty of the mining algorithm. Based on this, a model was created.
Eday is the cost of mining per unit of mining power per day, and M is the number of coins mined per day per unit of mining power. This model explained 92% of the historical price of bitcoin. Improvements have been made on this model to expand Hayes’s work by DeMeo and Young.
This model, in theory, could be used to value all minable digital assets. However, there is no way to tell how this will function looking forward as many cryptocurrencies move to eliminating miners (i.e. through Proof of Stake). Additionally, this model assumes that a change in the cost of mining or coins received would have an impact on the price directly — there are likely other key effects in between that would change the end result.
Given the current nature of the cryptocurrency market, absolute valuation methods don’t yield the useful findings by themselves. The real benefit when using them comes in combination with relative valuation methods. Since we can’t use P/E, EV/EBITDA, or the sort that we would in traditional equity markets, we must introduce some other relative valuation metrics.
We cover relative valuation models in depth in part II linked here
