Relative Cryptoasset Valuation
Every week the Mosaic research team will delve into important topics within the cryptoasset space.
By Jason Yannos & Lanre Ige
Introduction
A great deal of work has been done recently into intrinsic valuation methodologies for cryptoassets. For example, analysts have built on the work Chris Burniske has done with the equation of exchange to produce new models which aim to ascertain the intrinsic drivers of a token’s value. However, we believe there is value in novel relative valuation frameworks to allow intra-sector comparisons between similar cryptoassets. This article will take a macro snapshot of the current Decentralized Exchange (DEX) and ‘token burn’ token landscape and apply a relative valuation methodology to them. While the analysis is only a snapshot, it will act as the basis for further investigation.
Relative valuation methods
Often heard on the street is how institutional capital is looking to enter the world of cryptoassets. While the technology and scope of these projects is still emerging/nascent, investors face many challenges surrounding adequate valuation methodologies for these assets. Where on the investing spectrum do these assets fall? Should they be considered venture investments or should they be treated like investments in publicly traded equities markets with metric driven frameworks for assessing fair market values? The truth is there isn’t a one size fits all approach for any two cryptoassets given the various factors relating to each individual cryptoasset.
While much work has been done on the absolute valuation of cryptoassets, less work has been done on relative valuation — which estimates the value of an asset by looking at the pricing of comparable assets relative to common variables like cash flows, earnings, and sales. In relative equity valuation, multiples like Price to Earnings, Earnings per Share, Price to Book, and Dividend Yield are popular examples of multiples that are often used when determining the relative valuation of a company against another similar company or group of companies.
One of the reasons little progress has been made on relative valuation models is because of the esoteric nature of many of these assets and lack of comparable data points. In spite of this, we believe there is much to be gained from these frameworks, and that within the universe of cryptoassets there are many sectors and sub sectors that can be used for relative valuation.
One of the most promising categories for relative valuation is the decentralized exchange (DEX) space. The DEX Universe is comprised of a few major players — 0x, AirSwap, The Kyber Network, and possibly Bancor to name a few — alongside many minor players. All these protocols are not only live but provide comparable data alongside traditional on-chain data that can be used for fundamental analysis. For the purpose of our analysis of relative valuation in the DEX universe we will walk through a hypothetical multiple, referred to as the the Trades per User Ratio, and consider how this ratio may potentially imply a potential discount or premium for a token based on the protocol’s velocity.
Trades per User Ratio:
We collected data for the total number of users and trades for AirSwap, The Kyber Network, and Bancor and ran it through our ratio¹:
Similar to how an investor in a public traded equity might purchase or sell a company based off of a high or low P/E ratio, a cryptoasset investor could potentially make a buy or sell decision based off of the Trades per User Ratio. A higher Trades per User Ratio would imply that the tokens traded on the network have high velocity while a low ratio would imply that they have low velocity, comparatively showcasing how much activity actually take place. Although it would be naive to make an investment decision from one multiple only, a multiple like the Trades per User Ratio is simply an example of the types of ratios that can start to be developed in the DEX space to begin to form a holistic view on a specific token and the space as a whole.
Price to Burn Ratio:
One commonly touted value proposition for certain tokens is that of a ‘token burn’. This is a mechanism where a token is bought and then removed from the circulating supply — through the use of a burn() function or simply sending it to an address with a (likely) unknown private key. This decreases the supply, and for large amounts, it is argued, makes the remaining circulating tokens more valuable — assuming that demand for the underlying digital resource the token provisions remains constant (or increases) and that demand for the token is not extremely elastic.
There are similarities between a company buying back some of its publicly traded stock and the ‘buy and burn’ mechanism used by MKR and BNB. Whether the token burn mechanism is a defensible or ideal token mechanism is beside the point since it can serve as a fundamental metric by which to judge similar token projects.
We have created a model for what we call the “Price to Token Burn’ (PTB) ratio. The ratio compares the price of Binance Coin (BNB) and MakerDAO (MKR) to the dollar amount of their token burns. Both MKR and BNB use token burns to reward token holders for governance of the protocol (through the governance/stability fee) and to distribute value from Binance, the Company, to token holders who helped fund the company’s development. While the two token projects are extremely different, they offer a good chance to test out the PTB metric since both tokens are live and functional, with some amount of utility, as well as widely available data.
The Token Burn Model
It can be argued that a comparison of BNB to MKR on the grounds of their token burn mechanism is an apple to oranges comparison, and to an extent it is. Consider how the two token burn mechanisms work:
Binance
Every quarter, Binance uses 20% of its profits to buy back BNB and then destroy them; they will continue to do this until 50% (100MM) of the total BNB supply is bought back. All buy back transactions are publicly verifiable on the Ethereum blockchain.
MakerDAO
To create the Dai stablecoin, an investor sends Ether (or in the future other crypto assets) to a Collateralized Debt Position (CDP) on the Maker platform. When the investor wants to retrieve their collateralized Ether, they must repay the same amount of Dai they created, as well as a stability fee. The stability fee is denominated in Dai but the Maker system then buys MKR and burns it. All the steps above are publicly verifiable on the Ethereum blockchain. Visit this page to see a more detailed explanation of how the stability fee and burn mechanisms work.
The Dai system has only been active since December 17th, 2017 and many of the CDPs have not yet been closed nor had their stability fees paid.
For the purpose of this exercise:
- We assumed the yearly supply of Dai would be 100 million (the current single-collateral Dai debt ceiling) and the governance fee would remain at 0.5%.It is difficult to say how accurate these assumptions are, since the Maker team is likely to roll out multi-collateral Dai in the coming months — which will increase the total debt ceiling of the system further.
- Moreover, there is always the possibility that MKR holders will vote for an increased or decreased governance fee.
- The total stability fee amount accrued under our assumption would be 500,000 DAI.
- We used the current circulating supply given by the Dai dashboard.
From the above information, we can calculate the ‘Burn per token’ and the Price to Burn multiple follows:
- Binance burns tokens once every 3 months, but have only had three token burns thus far (the next will likely be in the next few weeks).
- We averaged the last three token burns to estimate the token burn amount (in BNB) for Q2 2018.
- With estimates of 4 quarters of token burns, we could calculate the total burn amount for a year and then the ‘Burn per token’.
The Price to Burn multiple follows from there.
Assumptions
Keep in mind that on a high-level, valuation multiples are comparisons of market value (price) to a metric which one thinks relates to the fundamental value of the asset. In traditional finance, earnings, the fundamental metric — cash flows, for example — used in the valuation multiple often serves to drive the price of the asset in question. For a valuation multiple to make sense for cryptoassets, there must be a clear logical relationship between a given cryptoasset’s price and the metric in question. This is difficult for most cryptoassets since their value propositions are often unclear and there are no clearly defined, tried-and-tested market value mechanisms for the space at the moment.
The token burn mechanism seemed like a plausible metric to use since there is an easy logical explanation for why reducing the supply of an asset, all things being equal, will cause the remaining supply to increase in value proportionally. However, both Binance and MakerDAO are new projects and the token burn functions have only been called a small amount of times for both projects. As such, the model was made with a exceedingly small sample size and it wasn’t possible to analyze the cross-sectional distribution of the Price to Burn metric. We suggest seeing the model as a thought experiment into what may be possible with specific relative valuation metrics for cryptoassets.
Here are some of the assumptions we made:
- We assumed that tokens are burned constantly throughout the year. Both MKR and BNB handle token burns differently.
- We assumed that the fundamental value proposition of MKR and BNB is the token burn. MKR holders may argue that ‘right-to-govern’ is the main value proposition of the coin. BNB holders may argue that the exchange fees discount or future expectations of the Binance decentralized exchange blockchain are the main value proposition.
- We assumed a full year of token burning for both projects and estimated future token burn amounts on that basis.
- We did not account for the liquidation penalty which, for single-collateral Dai, benefits PETH owners but a similar auction model will compensate MKR holders for the multi-collateral Dai launch.
Insights
Under the Price to Burn multiple, MKR is valued at a 58x premium to that of BNB. It could therefore be argued that MKR is overvalued solely based off its token burn function when compared to Binance. There are, however, two other other obvious explanations for this:
- People are valuing BNB and MKR for very different reasons. MKR tokens give holders governance rights of the decentralized Dai system, whilst BNB gains its utility simply from the Binance profit burn and exchange fee discounts. The Price to Burn multiple can be argued to not fully capture the value propositions of the two tokens. Ideally, this article would’ve compared tokens within the same category, with similar utility propositions; we were limited however by the lack of feasible tokens to analyze. Using two tokens with similar burn() functions is not necessarily sufficient to justify a valuation comparison.
- MKR is valued on expectations of future token burn amounts. The total addressable market of a stablecoin like Dai is potentially in the trillions of dollars. If we assume Dai scales to $1 trillion then, at the current stability fee, $5 billion worth of MKR would be burned annually.
Limitations
This article has studied an extremely small sample of tokens and no analysis was done on the cross-sectional distribution of the token burn multiple. The reason for this is that there are very few live projects which implement a regular token burn and the data that does exist is hard to find.
This analysis has also shown that using a similar mechanism — the burn() function — on different tokens affects the said token’s value proposition in markedly different ways. This fact makes it harder to develop useful, yet specific, relative valuation metrics. Despite these difficulties, we suggest it is important to continue making efforts to develop relative valuation models.
Conclusion
This article has offered two potential relative valuations metrics for DEXs and token-burn tokens. There is significant value in developing sector-specific metrics for cryptoassets given the large variance in token models. While more in-depth work in this area is currently bottlenecked by a lack of easily accessible on-chain data, we believe this will no longer be a problem in future. We envision a suite of relative valuation metrics for different token categories, such as Price to average amount staked (for Work tokens), or Price to cash flow generated (for cash flow generating tokens like REP). These relative valuation frameworks can be unique to different categories of tokens and, ultimately, it is crucial that investors are able to disambiguate between different types of tokens, and how their value accretion mechanisms differ.
This document is intended for informational purposes only. The views expressed in this document are not and should not be construed as investment advice or recommendations. Recipients of this document should do their own due diligence, considering their specific financial circumstances, investment objectives and risk tolerance (which are not considered in this document) before investing. This document is not an offer, nor the solicitation of an offer, to buy or sell any of the assets mentioned herein.
_____
[1] Data for 0x will be added shortly.
End of Biweekly Research Report
We hope you have enjoyed reading the Mosaic researchers’ biweekly report and stay tuned to the next article. Click here to visit our website
This article is intended for informational purposes only. The views expressed herein are not and should not be construed as legal or investment advice or recommendations. Recipients of this article should do their own due diligence, considering their specific financial circumstances, investment objectives, and risk tolerance before investing. The individuals contributing to this article have positions in some or all of the assets discussed. This article is neither an offer, nor the solicitation of an offer, to buy or sell any of the assets mentioned herein.
