Cliff wall

Why cliffs converge without anyone deciding they should

Vesting clocks start at the token generation event. If the investor bucket, the team bucket and an ecosystem grant bucket all carry a twelve month cliff, all three break on the same anniversary. Nobody chose that. It falls out of using the same start date and the same convention three times.

The near misses are worse than the exact matches, because they are harder to see. A nine month investor cliff, a ten month grant cliff and a twelve month team cliff produce three separate dates on three separate term sheets and one ninety day window carrying all three first releases.

And the first release from a cliff is not a normal monthly tranche. Most schedules bank accrual during the locked period, so the cliff date releases everything that would have vested in that time. On a four year vest with a one year cliff, that is a quarter of the bucket in one event.

Arbitrum, 16 March 2024, and what a single cliff can carry

Arbitrum's team, contributor and investor tokens vest over four years from 16 March 2023, with the first unlock exactly twelve months later, and the Foundation's own tokens vest over four years from 17 April.¹ Three insider categories on one date, with a fourth category one month behind it.

The 16 March 2024 event released 1,111,750,000 ARB, about $1.24 billion at the prevailing price of $1.12, equal to 87 percent of the token's then-circulating supply of 1.275 billion.² That is the shape of the problem stated in public numbers: one date, several buckets, and roughly a doubling of tradable supply.

Everything about it was disclosed a year ahead. The terms were published, the date sat on every unlock calendar.³ The size was arithmetic. Convergence being visible does not make it harmless. It makes it preventable.

We do not have a benchmark for how large a cliff wall gets

You will find pages quoting a percentage for how much circulating supply a cliff wall adds in a single month. We are not going to give you one, because no cross-project dataset supports it. The figures in circulation are repeated between blogs without a source underneath them, and we would rather leave a gap than fill it with a number we cannot stand behind.

What can be cited is named, dated events, and they vary enormously by float. The Arbitrum event at 87 percent of circulating supply is a documented data point.² It is one data point, and reading an average off it would be exactly the error this section is objecting to.

For your own project the number is knowable, which is the useful part. Build the month-by-month model across every bucket and the size of your own cliff wall stops being an industry statistic and becomes a figure you can read off a table.

Staggering costs a conversation and nothing else

Put every bucket's cliff end date on one calendar before any term is executed. Where two or more land inside sixty days, move one. Shifting an ecosystem bucket's cliff from twelve months to fifteen is a line edit during negotiation and a renegotiation with every counterparty afterwards.

Where a cliff length genuinely cannot move, the other levers still work. Release the cliff tranche in two or three parts across a quarter. Break accrual during the cliff so nothing banks behind the gate. Lengthen the vest that follows so the monthly rate after the cliff is smaller.

Most of the cliff walls we see in diligence came from teams that never built the combined view, not from teams that built it and accepted the risk. The aggregate is invisible until somebody deliberately assembles it, and assembling it is the cheapest work in the whole design.

What it does to a market, stated carefully

A cliff wall changes capacity, not intent. Several buckets become sellable at once, the market cannot distinguish which holders plan to sell, and it prices the whole overhang. Because the dates are public, that repricing frequently begins before the wall rather than on it.

We are not going to tell you what any specific token's price will do, and nothing here is a recommendation to buy, sell or hold anything. What we will say is that concentration is a choice made at term-sheet time, and spreading the same total supply across more days is available for free right up until the agreements are signed.