A Framework for Valuing MKR

I’ve been working on a basic discounted cash flow model to provide a high level estimate of MKR’s value and wanted to share it with the community – both to receive feedback and improve the model, but also to provide a lens into valuation for newer and less experienced investors. Shoutout to @Sebventures for providing valuable feedback and advice.

Anyone who’s been in the crypto community for more than a few years understands the immense amount of sh!tcoins out there with questionable use-cases and poor/obfuscated tokenomics. As we know, MKR is different – the value proposition and use-case is clear, fundamentals continue to improve, and the tokenomics are strong.

In this model there are several key inputs which drive the ‘fair’ value of MKR, all quite subjective this early in Maker’s growth. The discount rate and interest spread (difference between the long term interest rate on risk weighted assets and the DSR) are the two most important inputs. For most venture capital investments, a discount rate between 30-70% is often used. Typically the earlier one is in the investment cycle, the higher the discount rate used. If Maker was a traditional company, the discount rate would be lower, but in crypto there are unique risks – smart contract, legal/regulatory, central to decentralization, etc. Therefore, a higher discount rate was chosen both in the short/medium term and for terminal value calculations.

Regarding the interest spread, the ~5.5% rates we enjoy today will likely decrease over time as discussed in Seb’s Crypto Banking 101 post Crypto Banking 101. A 2% rate was chosen to be conservative though rates could stay elevated in the near term. Other key inputs include: YoY growth, Net Margin, and P/E Multiple. I’d encourage anyone interested in the model to adjust the inputs and enter their own assumptions to see how that changes the present value.



Good work @Aes. I have been making some DCF models for Maker in the past so I’ll make few suggestions:

  • rate spread and discount rate seem fine to me and I’d rather put more effort in simulating growth and breakdown of net margin.
  • linear growth seems a bit high to me (doubling each year), considering debt growth is still likely to be related to crypto market cycles (until RWA have larger share) and we are probably late in the current cycle. Simulations for growth as you have for spread and discount rate wouldn’t hurt here.
  • using fixed net margin seems too simplified, I’d imagine part of costs at Maker are going to be less related to growth than others. But it also isn’t an easy job to address this.
  • 30% of revenues that cover future losses, workforce (budgets + vesting), oracles could seem a bit too low, at least early on? I think recent budget simulations showed that 30% or more could be spent only by workforce and MKR vesting according to recent proposals?
  • I am missing liquidation penalties, are they part of rate spread? If so, spread is probably too low then.
  • I am also missing expected losses, but you could say that you simply covered them with liquidation penalties that you omitted? If so, these assumptions would probably need to be mentioned.
  • finally the big unknown, 10% of MKR held by foundation. We don’t know what happens with it, but I imagine it has a positive effect on your simulation either way.
1 Like

It’s column K. Must also say that we are saying we are late in the crypto cycle for months now (which I still believe).

In 2030, it’s $0.5T of deposit, this is less than half of Citigroup or Wells Fargo. A bit above Capital one which is quite recent (26 y.o.).

That’s a decision for MKR holders. I’m a big fan of targeting expenses using a normalized metric of net revenues. 30% on a $247M fee-base should make sense.

I also want to highlight that to keep DAI protection the same, we need to allocate a significant portion of the revenues when growth is significant. To avoid losing protection we need at least 40% of the SF to be used for that (limiting workforce expenses at 60% top, otherwise there is MKR holder dilution).

I’m on the side that you can’t issue half a trillion stablecoins without having a bit of regulation or at least appears safe to customers using regulatory standards. To reach 3.5% of leverage ratio, we need to allocate 70% of the net revenues to achieve it at the end of the decade.

With $MKR at 2k currently, there is a lot of price appreciation over the decade. It shows it’s possible. Not easy but possible. It’s an early model, it will be refined. But it shows that even the current price level magnitude depends on serious growth. We need to be focussed and dedicated on growth which is a bit my discussion about the business strategy (that currently we don’t really have).

Correcting some stuff, I found a current fair MKR price of 3.4k if we target 50% workforce expenses and 4.8k if we target a 70% workforce expenses.

PS: Regarding the excellent work, @Aes will be on the RWF Core Unit budget to continue improving the model (among other stuffs).


Agree @SebVentures, I personally also see protecting Maker with Surplus as priority number one and 1% capital/debt long term won’t work well. Just wanted to highlight that these recent Core Unit proposals (incl. upcoming ones with MKR vesting) might have an impact on it and we need to be aware of it.

One other angle of valuing MKR is that MKR value is not solely derived from cashflows but also retains governance value. In other words, you have MKR token holders who are interested in keeping DeFi secure and scalable through DAI because they have wider benefits in DeFi. So even if MKR was yielding 0, it would still have positive value for these reasons.

I think Coincenter had also one interesting “governance based token valuation” where it basically looks at the value governance token secures and then calculates token’s value through this bounty. So in MKR’s case it kind of serves as a floor value at 0 cashflow I think.


Thanks for the feedback! I’ve updated the model with a few key changes:

  • Broke out the Net Profit Margin by year to show higher near term expenses and margin expansion in latter years
  • Separated year end and average DAI outstanding for each year - DCFs now based on average DAI
  • Added assumption that expected losses would be offset by liquidation penalties
  • Revised some headers for clarity

I agree that the model growth is not realistic in the sense that we will likely continue to have boom/bust cycles around the halvenings until the crypto market matures. I’ve chosen not to model these cycles and smooth them out for simplicity though it is something certainly worth discussing.

Once we know what the foundation will be doing with the MKR I will update the model to reflect that but my assumption for now is that it will be held long term.


I agree 100% - next updates will be for the Surplus Buffer and Buyback to help illustrate the importance of protecting and maintaining an appropriate Surplus Buffer. I have also been working with Seb on tracking the CU budgets and proposals and update the Net Margin assumption once the first wave of proposals have been approved and we have guidance from the MKR Compensation Working Group.

I’ll have to follow up on Coincenter’s model - my initial thought would be that there is only Governance value if there are long term cash flows or some sort of value capture mechanism, but maybe I am misunderstanding?

I think the coincenter model can likely be applied to some of the higher cap tokens and the ones already used for governance. For others, with lower cap and short history, I personally would set up a backtesting approach based on a basket of crypto using for ex. ICO data and apply avg basket results across the collaterals.

This is an older interactive notebook attempting to perform a crude DCF on MKR. Obviously all DCF models require a leap of faith when the majority of the value comes from the end years.

One place I would push back is your discount rate of 40%. I understand this has become a standard discount rate among crypto analysts but it does not seem grounded in anything. If you believe that MKR is a venture style asset and venture capital has averaged a return of ~27%/yr for the past 30 years then 27% should be a starting top line discount rate. Unlike venture capital MKR is a liquid asset so it gets a liquidity premium over non liquid venture assets. Liquidity can be considered as an option. Researchers like Ibbotson have found this premium to be worth about 4% a year. So, 27% - 4% = 23%. One more thing to consider is that over the past 30 years the risk free rate was closer to 4.5%, today it’s 1.5%, so if you adjust the venture base risk premium by 3% it’s 24%. 24% - 4% = 20%, So 20% seems like a more realistic discount rate in today’s environment.


Thanks @bgits , very nice notebook.
Nice to play around with the sliders :stuck_out_tongue:

Note: even setting 50% probability of failure and 30% discount rate, the fair MKR price is computed as: ~8749k USD per 1MKR :slight_smile:

:clap: :clap: :clap:

1 Like

Thanks for your comments and joining the discussion @bgits. I think 20% would be appropriate for a traditional centralized firm and equity, but there are additional risks and factors to MKR and crypto which led me to decide on a higher discount rate.

These include regulatory risk, political risk (countries outright banning cryptocurrencies), cyclical nature of the crypto industry, smart contract risk, and the move from a centralized entity to a DAO. I would also note the higher willingness to take risk by governance as indicated in the decision to burn 25% of the increase of the Surplus Buffer.


I believe these are more risk factors for centralized entities. Decentralization removes certain risks that centralized entities have to deal with.


I think this is the biggest risk factor that warrants a large discount rate.

1 Like

The issue with regulation isn’t really with our core operation, it’s more with our real world assets side if we are decentralized governments can’t really touch us but they can prevent any legitimate business from receiving loans from our protocol.This would not kill maker obviously but it would provide us with significant hurdles in creating an uncorrelated basket of assets.

1 Like

This is a real risk, which is mitigated by two main factors at the moment. First, partnership with 3rd party ecosystems that assist the protocol doing business in a compliant matter with existing regulations, including AML/KYC and money transmission requirements. One partner here is Centrifuge which assists its asset originators in creating debt structures within the boundaries of what is permitted by each jurisdiction. The second factor, still under R&D, is the use of the PSM USDC for lending. USDC has been built to operate within the boundaries of regulatory environments with regulated entities. In theory, the protocol can provide lending facilities by deploying the USDC in the PSM to rwa AOs should they require it either by law or by necessity (volatility of DAI). But again, this use is under review at this stage.