Disclaimer: The author personally owns MKR. This content should not be considered investment advice.
One of the most respected projects in the space. Trading at near a Billion Dollar market cap. One of the few token models that, even token skeptics agree, is integral to the system functioning. Maker has achieved actual product market fit in a sea of userless projects. Yet we still seem to lack an understanding of how value accrual actually works in the MakerDAO system— and how much Maker (MKR) can stand to be worth in the long run.
In thispiece, I’ll outline a framework for understanding how we can value Maker in a mature market, and the potential & challenges that exist for MakerDAO.
Note: This is not an article for those new to MakerDao and a prior understanding of Multi-Collateral Dai (MCD) will benefit the reader. If you wish to learn more about MakerDao, please see the following resources:
When finding analogs for tokens in the traditional financial world, stocks are a commonly referenced asset class. This is especially true for tokens which have a burn model, as there is always a direct relationship between network activity and value of tokens burned — similar to how there is a direct relationship between profits and buybacks in the equities markets.
The resulting impact for token holders & shareholders is also similar when a corresponding token burn or equity buyback is incurred. A stock buyback increases each outstanding shares’ proportional ownership of the total company value while a token burn increases each token’s proportional ownership of the total network value.
A discounted cash flow (DCF) analysis is the most commonly used method to determine the value of the company. This valuation method is used because it allows an investor to determine the net present value of all future cash flows. This is the commonly accepted definition, but what does that actually mean? A typical shareholder in a publicly traded company is usually not actually entitled to directly receiving the entirety of the cash flows that a company produces.
Let's think about this in a more tangible perspective. Consider a scenario where an investor were to buy all the shares of a company — they would be entitled to all current and future profits of that company. Now if that investor chose to, they would actually be able to directly shift those profits from that company’s balance sheet to their own. This foreseeable stream of income is how a company is able to derive its fundamental value.
Here’s where the analogy starts to diverge. The Maker system does not generate profits in the same way a company does. So where does Maker derive its value? To answer this question, we need to understand the role of MKR in the MakerDAO system.
At a high level, MKR in Multi-Collateral Dai (MCD) will serve in the following capacities:
- Votes that can be cast in the proposal system
- A recapitalization resource during CDP liquidations (MKR is both printed & burned)
- The exclusive asset used to pay stability fees
The last two roles necessitate a clear continuous market demand for MKR in a functioning MakerDAO system. As liquidations continue to occur, and stability fees are continually levied upon CDP holders, MKR will be used and demanded from the market as an economic stabilizer. It is with this understanding that we begin to see that MKR may actually look more like a different asset class — a commodity.
In addition to having the characteristics, MKR certainly fits the classical definition of a commodity.
“An economic good or service that has full or substantial fungibility”
At first thought, it does not seem anything like the commodities that we traditionally think of when we hear the word — Gold, oil, wheat, sugar. But like these traditional commodities, for a certain market condition, there is a corresponding deterministic market demand and market supply.
A quick Econ 101 refresher — the quantity demanded and quantity supplied will differ based on the price point, creating what is known as demand and supply curves. The point at which these curves intersect is the market price — this framework can similarly be applied to MKR.
The above graph is a representation of a basic supply and demand graph. Let’s examine how the supply and demand curves for MKR might look like.
The immediate assumption some might make is that MKR has an almost completely inelastic supply — 1 Million Tokens were generated by MakerDAO, and the token supply is expected to slowly reduce over time through burning mechanisms. However, in the context of a practical market, the supply (i.e. tokens provided to the market) is actually elastic. At a certain point in time, there will be more people willing to sell their MKR At a price of $10,000/MKR than at $10/MKR. We can thus infer that the supply schedule for MKR is up and to the right, potentially resembling a linear curve.
As stated earlier, the utility functions of MKR in the MakerDAO system cause there to be an inherent & deterministic demand for MKR. The MKR that is demanded for stability fees & liquidations is consumed permanently in the same manner as conventional consumer commodities. Given the following annual average variables, a certain annual MKR demand denominated in Dai can be calculated:
- Total Dai Supply
- MKR Stability Fee
- Dai Savings Rate (DSR)
- Liquidation Penalties
These factors contribute to what we call the “Dai burn ratio” because the MKR demanded for stability fees & liquidation penalties are burned as part of the MakerDAO economic stability process.
A few notes about these variables in MCD:
- The DSR is a return for those that “lock” dai and is funded by the stability fee which results in less MKR being burned
- Liquidation penalties currently benefit PETH holders in SCD, but will benefit MKR holders in MCD
The dai burn ratio is the percentage of Dai’s total network value that is “paid” to the MakerDAO system through MKR burning on an annual basis. Applying the Dai burn ratio to the total DAI supply yields the total MKR demanded/burned in a given year:
Note that To determine the amount of native MKR demanded by the market requires an additional variable to be applied — the MKR/DAI exchange rate. The interesting phenomenon is that the higher the MKR price, the less MKR is actually demanded, and the lower the MKR price, the more MKR is demanded. The Maker demand curve follows a textbook power law distribution.
Unit of Account (UoA)
Typically for commodities, the quantity on the X-axis is measured in the units of that commodity. This seems to be self-evident — The quantity of crude oil is measured in barrels of crude oil, the quantity of corn is measured in bushels of corn, etc. This makes sense because the amount of utility a purchaser can derive from a commodity is directly tied to the units purchased of that commodity.
In the case of MKR, the amount of utility a user derives is not actually directly tied to a specific amount of MKR, but instead to an amount of MKR that translated from a certain DAI amount. Consider a case where a user takes out a 10,000 Dai loan with an average 3% stability fee. After a year, he will need to pay 300 Dai in MKR in order to close out the CDP. The exact amount of MKR he will need to pay will vary based on what the price of MKR is at the time of payment, but the users cost in Dai is consistent regardless of the MKR/Dai exchange rate.
From a user perspective, it is most practical to quantify MKR in terms of Dai.
With this understanding, we can determine that the Demand Curve is inelastic for a given stability fee & Dai supply.
When the unit of account changes, not only is the demand curve transformed, but the supply curve is as well.
Let’s examine what happens when the price of MKR increases:
- The total amount of MKR (denominated in Dai) available for fee payment increases i.e. The market cap of MKR doubles when the price of MKR doubles
- More people are willing to sell their MKR
The impact of (1) is a predictable function, however, the impact of (2) is a relatively complex function determined by a multitude of factors (e.g. risk tolerance, willingness to sell, etc.) that characterize the supply side market participants. If we assume that (2) is a linear function, then the supply and demand curve would appear as follows:
MKR Supply & Demand
The supply curve follows a root function: Each increase in MKR price results in a subsequently larger amount of MKR (denominated in Dai) available to the market supply.
Given this change in perspective for the supply & demand UoA, one also realizes that Maker has an (almost) infinite supply as a commodity product. While the total number of MKR is capped at 1,000,000, the price of MKR, and therefore the value of MKR denominated in Dai is unbounded. The amount of Dai supply can reach infinity, and the MKR supply required to pay the fees in the system can expand accordingly, also to infinity. This same property combined with extreme divisibility (to 18 decimal places) is what allows MKR supply to meet demand regardless of the actual number of MKR coins.
With this understanding of MKR Supply & Demand dynamics, we now have the foundation to formulate a valuation framework. This starts with being able to approximate the value of MKR demanded/burned in a given year. This value can be interpreted as not value that is permanently destroyed, but rather the amount of value that is distributed to the non-burned tokens.
If the MKR demand in the subsequent year is the same as in the prior year, then the supply of MKR (denominated in Dai), must counteract the burning of MKR to meet that demand. As explained previously, the mechanism through which supply expands is through price increases, thereby benefiting MKR holders.
For example, consider a situation with 1 Billion Dai Supply & a 3% Dai Burn Ratio with a MKR market cap of $600 Million. At the end of the year, 30 Million Dai ($30M) worth of MKR will have been burned (technically some will be earmarked to be burned as some CDP stability fees maybe unrealized). If the Dai Supply stays constant, the MKR demand for the subsequent year will still be $30M. If all other factors are constant as well then MKR holders willingness to sell at each price level changes i.e. the supply curve shifts to the left. Consider the perspective of a rational MKR holder — if one holder owned 1% of MKR supply ($6M worth), then post burn, he would now own 1.05% of the supply as 5% ($30M/$600M) of the MKR supply was burned. That rational MKR holder would expect their principal ownership % (1%) to entitle them to the same value of $6M both pre-burn and post-burn. However, because 1% of MKR supply represents a smaller amount of tokens post-burn, the dollar value per token must increase proportionally.
Over time, the MKR demand curve will be shifting to the right — assuming a Dai supply growth that outpaces any decreases to the Dai Burn Ratio. On the other hand, as more MKR is burned, the MKR supply curve will shift to the left. Both of these trends contribute to upward pressure on MKR price.
The percentage of MKR supply that is burned on an annual basis is referred to as the MKR Burn Rate. Given the value transference dynamic, it can also be thought of as an inherent rate of return. If we can calculate a rate of return, then we can apply the standard principles of cost of capital. In essence, the rate of return will equilibrate to a level that investors are willing to accept given their expectations of risk & future growth for a MKR investment.
This is traditionally how investors think about investments —higher risk investments (i.e. stocks) necessitate a higher return, while lower risk investments (i.e. US Treasuries) will be bid down to lower yields by the market given their perceived risk level compared to other investments.
In the case of MKR, the higher the MKR market cap, the lower the burn rate/rate of return, and vice versa. With this perspective, we can model out potential MKR market caps that correspond to various rate of returns in different scenarios.
Let’s look at the scenario in which Dai Total Supply is at 1 Billion Dai. At a 3% Dai Burn Ratio and a 15% Annual Return, the market would price MKR at a $200M Market Cap. Compare this to 1 Billion Dai and a 3% annual return — the market would price MKR at a $1B Market Cap. In other words, the more comfortable investors are with the expected burning/demand of MKR, the lower the expected return they are willing to accept — resulting in the price/market cap being bid up to a higher amount.
Using this framework, we can come up with up a (rough) estimate of a valuation that MKR may potentially achieve. To arrive at this estimate, there are three variables that are important to focus on:
- Dai Supply
- Dai Burn Ratio
- Expected Annual Return
The total Dai supply is a function of both the collateralization ratio and the assets collateralized.
The average collateralization ratio currently hovers between 250–300%. For my base case I assume a collateralization ratio of 150%. As more lower volatility collateral types (i.e. tokenized assets) are added to MakerDAO in MCD, the liquidation ratio, and average collateralization rate will fall. As a point of comparison, Bitcoin/Ether are currently ~10x more volatile than equities, and ~30x more volatile than global treasuries.
Not only will less volatile assets be available for collateralization, but native cryptoassets will also reach a more mature market status causing their own volatility to drop.
We can bucket the potential assets to be collateralized into 3 main categories:
- Permissionless Cryptoassets
- Tokenized Fiat
- (Semi)Fungible Non-Bearer Assets (e.g. stocks, treasuries, corporate bonds, commodities)
Currently, the only asset available for collateralization in MakerDAO is ETH, a permissionless cryptoasset. As of the time of writing, MakorDao currently has 2.13M ETH locked as collateral — this translates to ~$300M. The upside potential is much greater.
Multicoin Capital estimates the market opportunity for permissionless cryptocurrencies to be worth $50–100 Trillion. If we take a low-end estimate of $50 Trillion, a 2% capture rate by MakerDAO means $1T in assets collateralized. A 5% capture rate means $2.5T in assets collateralized. 20% means $10T. For context, in just over a year of launching, MakerDAO has already locked >2% of total ETH supply.
There are a couple of factors that would imply a higher or lower percentage of permissionless cryptocurrencies collateralized. One is that the current ETH collateralization rate is based on Maker’s progress in its infancy and stands to grow at a continuoly fast rate over time. On the otherhand, Maker has the momentary benefit of lack of competition — CDPs are currently the best tool to leverage ETH —decentralized alternatives like Dharma & centralized alternatives ala BitMex will likely become larger players in providing ETH leverage.
Tokenized fiat represents another large pool of potential collateral. A commonly cited scenario is that M1 and parts of M2 of global money supply will eventually be tokenized. This represents a potential asset base of $40–100 Trillion.
Now let’s consider semifungible non-bearer assets. It is estimated that there are >$50 Trillion in outstanding global treasury debt. Global treasuries are one of the most liquid and least volatile assets, and would serve as prime collateral in MakerDAO.
Now take into account other high potential tokenized assets that could be easily used as collateral in MCD — global equities, corporate bonds, commodities, etc.. The total addressable asset base grows to $300T+. Depending on one’s assumptions for the collateralization potential of these assets, there is a clear path to $1.5T assets collateralized, and potentially $10T+. These numbers correspond respectively to a $1T total dai supply and a $6.7T dai supply.
It is important to note that permissioned assets like tokenized fiat & non-bearer assets represent a systemic risk to MakerDAO. They have the potential to be frozen by government entities or organizations that manage the smart contract tied to these tokens. However, MakerDAO has stated that they plan to add these assets and have plans to minimize these risks.
Dai Burn Ratio
As stated previously, The Dai Burn Ratio is made up of three components — the liquidation penalities, stability fee, and DSR.
Currently, liquidations make up a significant component of all value accrual mechanics in the MakerDAO system — Marc-André Dumas calculated the exact number to be 68%.
However, with the introduction of secondary market insurance products (one of which is almost live) & lower volatility of collateral assets, I expect liquidations to be marginal and trend towards 0, or an insignificant quantity, in the future.
The stability fee is basically the interest fee that CDP owners pay for their loan. The interest rates of secured loans work as a proxy, as they are similarly backed by assets. Looking specifically at the 30-year Fixed Rate Mortage, we find that the average rate floats between 4-7%.
Estimating an expected value for the DSR is a bit tricky. The DSR serves as a key driver in increasing demand for Dai such that it can maintain it’s $1.00 peg. DSR has not yet been implemented, so we don’t have any historical data to inform what a range of rates may look like in the future. However, what we have seen is that a low stability fee combined with a lack of DSR results in a supply & demand imbalance as evidenced by MakerDAO’s recent proposal for a stability fee increase. My estimated range for DSR is 2–5%.
Taking this all together, we can approximate a 2–4% Dai Burn Ratio. Here are a few scenarios modeled out:
Expected Annual Return
Now for the final question: What rate of return will investors be willing to accept?
Their are a few factors that influence the rate of return investors will accept:
- Expectation of stability fee & DSR changes
- Expectation of contraction OR growth of Dai supply
- Risk of smart contract bug/ other system failure
- Potential returns of other investments
Depending on market conditions, it is likely that the rate of return ranges between 3–6% in the long run after Maker is a proven system & model — similar to the returns of other debt instruments.
Note that the current burn rate from stability fees is quite low — ranging from 0.1%-0.6%. This is because MKR is trading at a premium representing investors expectations for its future growth.
An interesting peculiarity of this approach is that you arrive at the same outcomes as you would with a DCF on burn rate approach. The rationale behind the DCF approach is similar to the approach we reviewed — burn rate is interpreted as a return for MKR holders.
In fact, if we return to the analogy of a single investor purchasing the entirety of the outstanding shares of a company, we find we can extend this example to MKR as well. The difference is that instead of profits, an investor can expect a certain dollar amount of MKR they can sell on an annual basis. Even though this investor would own the entire supply, and have full pricing power on a MKR-USD basis, the demand inelasticity puts both a floor and ceiling on the amount of dollar amount of sales a controlling investor would be able to generate — whether that investor prices his MKR at $100,000/token or $100/token.
Note that in this framework, Governance, although a non-marginal value driver, is not included and deserves a thorough analysis of its value fundamentals itself.
Taking this all together, while we do not know exactly how the multitude of variables will playout for MakerDAO, we can use informed assumptions to determine the potential value for MKR. I will leave it up to other analysts to make their own assumptions around the variables outlined, but it is clear that there are many potential scenarios which lead to a Trillion Dollar value.
Maker’s widespread adoption as a global credit facility will not happen overnight, in fact, it will likely take decades if it happens.
Nonetheless, I maybe even understating the potential value capture of MKR when evaluating it only in the context of a credit facility:
Thanks to Nik Kunkel, Aleks Larsen, Jason Choi, and many others for long discussions around everything MakerDAO that have shaped my thoughts on this.