Alex,

I want to congratulate you for the excellent post. I also enjoyed your mechanism design post a lot — really good stuff! It is clear that you’ve put a lot of thought and effort into it! I completely agree with your thesis and the velocity issues. Funnily, exactly two days before you published this, I took a stab at some of these issues, albeit from a slightly different perspective — focusing on discount rates, some additional velocity factors and also a flaw I found in some of the earliest valuation models regarding discounting. I would be glad if you let me know what you think about those: https://medium.com/@hristovbz/thoughts-on-token-valuation-dynamics-9ecb979b7b65

Let me come back to a few comments on your model:

1/ Cell B45 in the model says “Whalesale” instead of “wholesale”. I don’t know if it was intentional, but it is really funny and spot on :) People in crypto should adopt it immediately :)) Then cell G72 says “Retutrn” —just a typo. Also cell B12 — “Elecricity”, missing a T.

2/ In your transaction cost analysis, I would add this: many projects would initially add a fiat option in addition to the respective token option for transacting. This obviously makes a lot of sense from an UX point of view and user adoption. So I would also consider the simple option where a user purchases electricity with his credit card leading to the merchant paying the CC transaction fees like any other purchase. His money will be converted into the utility token internally by the project’s smart contract and then back to fiat at the other end to pay the electricity “whalesaler” :) New projects will have to keep an inventory of tokens for this purpose even before exchange listing and then also manage the price stability between exchanges and their own inventory (this could have very interesting dynamics). The point being, users may have no problem in transacting frequently with zero fees due to design decisions at the token level. If you put zero in place of the 20 USD, the model breaks down as the utility value basically nears zero. This is why staking design seems to be so crucial in order to be able to attract investors upfront.

3/ I discuss this extensively in my post, but there should be a much larger discussion around the “risk free rate” and discount rates. I would argue that lots of people in a post-crypto world keep their funds in BTC for example. You would probably have different ideas about risk-free rates then. Alternatively, with ETH, we could imagine a state where you keep your funds in ETH which inflates by say 3% a year and you get 3% risk-free rate by staking. Separately, the discount rates of 40% seem fine for an outside investor (outside meaning a cross asset investment fund) as they would add a 20% crypto premium on top of 20% required rate of return for VC as an asset class, but take a crypto only fund which has crypto only LPs and their discount rate may just include the specific token risk premium over BTC or ETH and hence be like 15–20% (who knows..).

4/ On velocity, I would also imagine that depending on the project, there would be stakers who will have to naturally hold tokens (take the PROPS model as an example) and those who will hop on and off the token based on the market vs. current utility vs. the discounted utility price, all depending on their time horizon (you are absolutely right about the utility value having a maximum at some point).

Take care

Best, Boris