The IC Launch: A market perspective
Note: Any assumptions made below are just educated guesses, exact figures are subject to change once the team releases more detailed info on token distribution.
The Internet Computer has always been an ambitious project. One that required years of intense research, prototyping, testing, re-writing, more research, more testing, and finally... launch.
No launch is perfect, try as you may. A project launch like this undergoes scrutiny from all kinds of participants - from early investors, to cryptocurrency traders, large market makers, developers from other communities in the industry, press, and of course, the masses on Twitter and Reddit. In this post, I'll share my understanding of launch from a market perspective.
On May 10 2021, 1600 UTC, ICP markets opened on Coinbase. Liquidity came from 4 primary sources:
Initial Liquidity from Finance
When a new asset launches on exchanges, centralized or decentralized, it must have enough liquidity for accurate price discovery to occur. For most projects, some of this is provided by the project itself (in this case, the finance team) using its treasury. DFINITY is no exception here, with up to 3.5M ICP being distributed to Coinbase, Binance, Huobi, and OKEx. We don't have any information on how the proceeds from selling this ICP were used, nor do we know what the market maker fees were.
These transactions are among the early batches distributed from Neuron 4000.
Potential Tax Sales
When a treasury comprised of illiquid tokens suddenly has a market for them, the increase in valuation usually comes with a tax burden. In order to pay for this, it is likely that tokens were sold. I am guessing that account 32c4, which was allocated 10M ICP, has been used for taxes. 1M ICP from this account was transferred to exchanges at launch.
Restricted DFN Agreement holders
Employees who joined prior to mid 2018 or so were offered a "Restricted DFN Agreement", similar to how traditional startups offer an options package. These options contracts could be exercised in exchange for future ICP tokens, and came with a very standard 4-year vesting/1-year cliff. This contract was slowly phased out in favor of "Restricted Token Units" (RTUs), which are closer to RSUs (Restricted Stock Units) and have different tax implications than options. RTUs are also on different vesting schedules, with the first vesting occurring on June 24, over a month after launch. Options holders had the unique benefit of being fully liquid (vested amounts) at network launch. Up to 20M ICP from this group made up the majority of day 1 liquidity, all being distributed from Neuron 4000.
Finally, we have the remaining participants. This could include other early employees and investors who had different arrangements or other nonstandard contracts. We are unlikely to know the exact details of these arrangements due to NDAs or other legal obligations. Some of the previous 20M liquidity could actually be held by this group.
Who was not a market participant at launch?
It is important to note that Genesis Account holders (Seed Round/Early Contributor) were required to undergo a KYC process in order to claim tokens. This, of course, is standard practice for compliant token sales nowadays, but was not back in 2017 and 2018. These participants were only able to start the KYC process right around network launch, and the time that the first participant could access liquidity was 2021-05-10 20:54 UTC, 4 hours after trading opened.
The following table summarizes when Genesis Accounts were able to first access liquidity.
NUM SUM_ICP DATE -------------------------- 1 1660.38 2021-05-10 4 329137.77 2021-05-11 11 100579.97 2021-05-12 6 185230.16 2021-05-13 2 10885.08 2021-05-14 29 189171.42 2021-05-15 9 45014.30 2021-05-16 5 18970.77 2021-05-17 22 236006.17 2021-05-18 9 32304.99 2021-05-19 24 135407.08 2021-05-20 2 9378.63 2021-05-21 5 14120.32 2021-05-22 1 2489.51 2021-05-23 13 80928.11 2021-05-24 19 70432.47 2021-05-25 7 21080.25 2021-05-26 2 6193.51 2021-05-27 1 4188.20 2021-05-31
In aggregate, only 880k ICP of liquidity from the Genesis Accounts was available in the first week!
Likely due to listing requirements imposed by Coinbase, the Foundation was unable to divest any of its own treasury holdings for the first week after launch. The Foundation incurs expenses (eg. payroll), which must be paid for in fiat currencies - any functional foundation must sell some of its holdings in order to operate. It is likely that the Foundation was only able to sell some of its treasury after May 24, at an average ICP price of less than $150.
We know that Presale and Strategic investors are on 12 and 36 month vesting schedules, respectively. However, we do not know if any participants had special commencement date or upfront liquidity clauses in their terms. Perhaps some of them did have access to at least part of their allocation on launch day.
Most Current Employees
As mentioned earlier, RTU agreements only started distribution on June 24, long after the market open. The vast majority of current employees are on this arrangement, with the transition from options to RTUs having been completed by June 2020. RTUs vest over 4 years, and come with tax withholding capabilities. There is also now an employee program that incentivizes long-term staking.
As is the norm for startups that grow very quickly, compensation packages for early employees who take a risk and join a new venture can skew quite high, before being standardized (and decreasing) over time. Allegations that "each employee owns $15M in liquid tokens" are simply untrue, and frankly, offensive.
Summary of Liquidity
We can estimate that up to 25M ICP of liquidity was available at launch. We can see that the vast majority, upwards of 81.6%, of this initial liquidity was provided by the "Restricted DFN Agreement" holders. Foundation provided liquidity for market making and tax purposes made up the remaining 18.4%.
We know that the total supply of ICP at launch was around 469M, and 25M is a little over 5% of supply. We also know that the ICP Futures market on FTX was popular before liquid ICP was available. The limited supply, combined with the sentiment-based speculation common with new token launches, created the conditions necessary for ICP to trade at up to $700. Over the next few weeks, supply from Genesis Accounts slowly became accessible, the first distributions for Presale and Strategic investors were sent, and the price gradually declined. The fact that the broader market sentiment was turning bearish only exacerbated the downtrend.
Clearly, this was not an ideal situation. What could have been improved?
Genesis account KYC should have started months before launch. Month 0 Neuron should have been fully unlocked and ready to be supplied to the markets the moment that trading opened. This would have allowed much better price discovery.
Genesis account holders needed better tooling at launch. There are always so many priorities to balance when launching a network. Unfortunately, tooling for seed investors was simply not high enough on that list. Having a smooth UX would have given these investors the confidence to actually move their funds and help alleviate supply issues.
Restricted DFN Agreements should have been less favorable to ex-employees. Employees who voluntarily left DFINITY (or were let go) kept these very favorable terms. Some higher level employees, such as VPs or Directors who joined in 2018, could have negotiated very generous compensation packages, then leave (or get fired) and have the ability to maximize their profits in the first days of trading. Of course, this agreement was legally binding, so there really was no way to address this issue.
We can all observe that the ICP price has declined by 95% since all-time high, but that is ignoring the fundamental issue at play. Launching a network is hard. It will never be perfect. Sometimes the market conditions dictate that you must launch now. In the rush to deploy so many moving parts, some things get neglected: token management UX and KYC processes among them.
It's easy to buy into false narratives and jump on a bandwagon. We hear accusations of insider trading, of misleading investors, and of the team wildly profiting from retail investors - all very much attention grabbing, but very much lacking in nuance. It is ridiculous to say the team was negligent or acting maliciously, but certainly, we can blame them for not executing the perfect launch.