A network token has one hard job: pay for security and growth until fees can carry part of that cost. Layer 1 tokenomics should therefore be judged on the year fees are expected to matter, and on who can change issuance if they do not. Solana, for example, started at 8% annual inflation, cutting it by 15% a year toward a 1.5% floor, and a stake-weighted governance vote in August 2026 approved doubling the rate of decline.
Who decides this
The foundation council owns the network’s token economics. It is accountable for three things holders cannot see in a token paper: whether security is paid for at a price the chain can sustain, whether fee revenue reaches the token or the treasury, and whether the rules can be changed by anyone who will still be in the room in year three. The core engineering team proposes parameters. The council should sign them off as a budget, not as code.
What should pay for security: issuance or fees?
At launch, issuance pays for security, because fees are too small to do it; the council’s decision is how fast issuance falls and what replaces it. Every Layer-1 makes this trade. High issuance pays validators well and dilutes everyone who does not stake. Low issuance protects holders and can leave validators underpaid before fee volume arrives.
Solana’s published schedule shows the standard design. Its inflation schedule begins at 8% a year, reduces that rate by 15% year on year, and settles at a fixed 1.5% long-term rate. The schedule is predictable, which is the point: stakers and exchanges can model it.
Ethereum chose a different lever on the fee side. Its London upgrade on August 5, 2021 introduced EIP-1559, which splits a transaction fee into a base fee and a priority fee and burns the base fee. The Block reported that about 32,100 ETH was burned in the first week after the upgrade, roughly half of the 61,600 ETH issued to miners over the same period. A burn does not pay validators. It returns value to all holders when usage rises, and it does nothing when usage falls.
Our view: set issuance against a security budget, not a target yield. Decide what it costs to keep enough stake at risk, fund that, and let fee capture do the rest.
In practice that means three numbers in the token paper. The first is the share of supply the council wants staked. The second is the yield needed to attract it. The third is the year in which fee revenue should cover a stated share of validator income. If the third number is missing, the schedule is a hope rather than a plan.
What does a Layer-2 token actually capture?
In the two Layer-2 designs below, the token captures no fee revenue directly; the revenue sits with the sequencer side, and the council must decide how it reaches the treasury and what the token is for. That makes L2 token economics a governance and treasury question more than a security one.
Arbitrum shows the mechanics. The ARB token launched with a 10 billion supply, and the DAO can mint up to 2% of supply a year through a constitutional proposal. The DAO’s revenue comes mainly from sequencer margin: what users pay, less the cost of posting data and settling on Ethereum.
That margin moves with protocol upgrades. In a June 2024 request for comment on sequencer revenue, Entropy Advisors found the DAO earned about 2,125 ETH in profit over one 71-day period and about 938 ETH over the next, a fall of more than 55%. Over the same early-2024 months the DAO spent an average of 13.8 million ARB a month, with more than 90% going to incentives.
Optimism took a network-level route. Under terms it described in August 2024, each chain under its Standard Rollup Charter pays the greater of 2.5% of chain revenue or 15% of onchain profit to the Optimism Collective. At that point the Collective reported more than 14,000 ETH generated to date. The same post said governance had not yet decided how to allocate that revenue.
For a council, the practical point is that the token’s case can rest on decisions governance has not yet made. Holders will ask what the token is for once sequencer revenue grows. The council should answer that before launch, even if the answer is a treasury under a published spending rule, rather than leave it to the first contested proposal.
| Dimension | Layer-1 | Layer-2 |
|---|---|---|
| What the token is for | Staking that secures consensus | Governance and treasury, not settlement |
| Main supply lever | Issuance schedule and any fee burn | Governance-controlled minting and treasury release |
| Where fee revenue lands | Validators, or burned | Sequencer operator, then treasury if governance routes it |
| Main cost line | Validator rewards | Incentives and grants paid from the treasury |
| Who can change it | Validators and stakers, through protocol governance | Token holders, through DAO proposals |
| Typical failure | Issuance falls before fees arrive | Treasury spends faster than sequencer margin replaces it |
Who can change the schedule, and how hard should it be?
Changing issuance should be possible but slow, with a supermajority and a public model; the council has to decide the threshold before launch, not during a price decline. Too easy, and the schedule is a promise anyone can break. Too hard, and a bad parameter stays in place for years.
Solana’s August 2026 vote shows both sides. SGP-0002, which mandates doubling the yearly disinflation rate from 15% to 30%, passed with 67.001% against a 66.667% threshold, on 60.7% participation of eligible stake. It was the network’s first binding on-chain governance vote. The vote is a mandate: client teams still have to ship the change and validators still have to activate it.
Optimism’s governance forum shows the Layer-2 version. In 2023, a governance proposal to set OP inflation to zero covered May 2023 to April 2024. At the default 2%, about 85.9 million OP would have been minted into the treasury on a 4,294,967,296 total supply. The design question is not the answer that year. It is that a yearly decision exists and someone has to bring it.
We would write three things into the launch documents: who may propose a change, the voting threshold, and the model every proposal must publish. Without the third, governance debates become arguments about yield rather than about the security budget.
How should the ecosystem allocation be spent?
The ecosystem allocation should be spent against measurable fee growth, because it can be the largest discretionary cost the council controls. Validator rewards are formula-driven. Incentives and grants are choices.
The Arbitrum figures above show why this matters. A treasury spending more than 13 million tokens a month against a sequencer profit measured in hundreds or low thousands of ETH a quarter depends on token price, not on revenue. That can be the right bet for a period. It should be a bet the council made on purpose, with a review date.
The council should ask for one number before approving any programme: expected fee revenue per token spent, and when it will be measured. Report that figure every quarter next to the issuance curve, so the council sees both cost lines on one page. Attracting the first projects that generate those fees is a separate decision, with its own owner in the ecosystem team.
What goes wrong: the strategy nobody could execute
The token paper assumes fee revenue will replace issuance by year three. Nobody is named to track that assumption, and no parameter is tied to it. When year three arrives, fees have not grown, issuance is still falling on schedule, and validators are leaving.
The council now needs a governance change. It has no agreed threshold, no published model and no owner. Proposals stall for months while stakers argue about yield. The strategy was sound on paper. It failed because nobody could execute the adjustment it always required.
The defence is to treat the token model as an operating plan. Name the person who reports fee revenue against the issuance curve every quarter. Agree the trigger that sends a change to governance. Publish the model that any proposal must use. Then test the process once, on a minor parameter, before it is needed for a major one.
How Alcyone approaches it
Alcyone works on the parts of a network’s token economics that have to be executed: the first launches on the chain, the vesting and lock structures behind them, and the reporting that follows. In one engagement described in Alcyone’s Layer-1 advisory case work, the first token launch on a major Layer-1 left about $10M locked in vesting from that launch. Every plan Alcyone builds names an owner for each part, and there is no obligation to use TrustSwap products.