Engineering · Tokenomics

Coinbase splitter live: 37.5% validator-coinbase burn sealed on chain 2800

By AERE Network May 31, 2026 5 min read

Correction, July 20, 2026. This post was originally published as "AERE goes deflationary" and that headline was wrong. Direct measurement since publication shows that validator coinbase revenue on chain 2800 is zero, for two independent reasons: genesis declares no blockreward, and the 1 Gwei base fee is destroyed by EIP-1559 rather than paid to the proposer. Because the 37.5% burn is a share of that reward, it burns effectively nothing. Lifetime burned is 0.1374 AERE against a 2,800,000,000 AERE supply. Aere is not deflationary today. The splitter and the burn vault are real, live, sealed and ownerless exactly as described below, and the mechanism will burn in proportion to real fee flow when there is real fee flow. The forward-looking passages in this post should be read as the intent they were, not as a description of what is happening now. We are correcting this in place rather than quietly deleting it.

On May 31, 2026 AERE Network deployed AereCoinbaseSplitter, a tiny atomic contract that finally backs the whitepaper §3.3 claim with on-chain reality. Validator coinbase rewards route through it on a sealed 37.5% to permanent burn, 62.5% to the validator split. In practice that reward is currently zero, so the split has processed 0.1374 AERE over the chain's lifetime. The burn vault has no withdraw function, no admin escape, no upgrade path. What goes in stays burned forever.

What this changes: The whitepaper has promised since launch a 37.5% burn (a Foundation-set rate, capped at 50%). Until today, that claim wasn't backed by on-chain mechanism. Now it is: AereCoinbaseSplitter routes 37.5% of every block's validator coinbase reward to the sealed burn vault, verifiable by anyone in 30 seconds via splitter.totalBurned() or the burn vault's balance on the explorer.

The contract

Contract	Address	What it does
AereCoinbaseSplitter	0xb4b0eCe9011613A5b84248a9B42a0f309E6F01Ec	Atomic 37.5% / 62.5% splitter. Validators call `splitToSelf()` (or `splitAndDistribute(validatorAddr)`) with their accumulated coinbase as `msg.value`. Contract atomically forwards `3750/10000` of value to AereFeeBurnVault, remainder back to the validator's address. Emits `Burned` and `Distributed` events with cumulative totals. Burn rate is capped at 50% by contract code; Foundation can adjust within that ceiling.
AereFeeBurnVault (already deployed)	0x696afDF4f814e6Fd6aa45CE14C498ed9375fB2c6	Permanent burn destination. Has no `withdraw` function. Has no admin. Has no upgrade path. Once AERE enters this contract, it is mathematically unreachable. Anyone can verify this by reading the bytecode.

How the flow works

Three layers, each transparent on chain:

Validators produce blocks normally. Block rewards (gas fees + emissions) accumulate in the validator's coinbase address as usual. No change to Besu, no change to QBFT, no hard fork required.
Each validator runs a small forwarder daemon. A ~80-line Python service watches the validator's coinbase balance. When it crosses a threshold (default 10 AERE), the daemon signs and sends a transaction that calls AereCoinbaseSplitter.splitToSelf() with the balance as msg.value.
The splitter atomically burns and returns. The contract sends 37.5% to the burn vault (permanent), 62.5% back to the validator's address. Net result: the validator keeps the majority of their earnings while the chain's deflation accrues automatically.

The forwarder daemon is the only off-chain component, and it's tiny: ~5 MB RAM, runs alongside the existing Besu node, with no new infrastructure required.

Why this design (and not a Besu fork)

The alternative, modifying Besu's block producer to enforce the split at consensus level, would have required forking Hyperledger Besu, maintaining the fork against every upstream release, and coordinating a hard-fork-style network upgrade for every parameter change. That's real engineering cost.

The splitter-plus-daemon design is observationally identical from outside the chain. The on-chain math is the same. The cumulative burn is the same. The only difference is that it ships in days instead of months, costs nothing ongoing, and leaves Besu unchanged.

The trade-off: validators technically could stop running the daemon and pocket 100% of fees. But there are 7 validators today, all Foundation-operated, and any divergence between coinbase receipts and splitter calls would be immediately visible on chain. Public accountability is the enforcement mechanism today. Phase 2 (post-audit, with hard-fork upgrade) can move this to consensus-level enforcement.

What's measurable now

Anyone can query in O(1):

AereCoinbaseSplitter.totalBurned(), lifetime AERE burned through this contract
AereCoinbaseSplitter.totalDistributed(), lifetime AERE returned to validators
AereCoinbaseSplitter.burnedBy(addr), per-validator burn contribution
AereFeeBurnVault.balance, total locked AERE in the vault
AereCoinbaseSplitter.burnBps(), current burn rate (3750 = 37.5%)

The network status page now displays a live "Deflation tracker" card showing the lifetime burn in real time. The number is 0.1374 AERE and it is effectively static, because the daemons can only forward what the validator coinbase actually receives, which today is nothing.

Tokenomics impact

Correction to this section. As originally written this paragraph claimed a "per-block emission schedule from genesis" and a "slowly-inflating supply curve". Both are wrong. The canonical genesis for chain 2800 declares no blockreward key at all, so there is no emission schedule and never was one. Supply has been fixed at 2,800,000,000 AERE since genesis-v2 and cannot increase. AERE was never gas-inflationary.

With the splitter active, the picture changes:

Actual flow per block today: the EIP-1559 base fee (1 Gwei) is destroyed by the protocol, and the priority fee is zero, measured as effectiveGasPrice equal to baseFeePerGas on every recent transaction. Validator coinbase therefore receives nothing.
What the splitter does with it: of whatever does reach the coinbase, 62.5% goes to the validator and 37.5% is permanently destroyed. That split is live, sealed and correct. It has processed 0.1374 AERE over the chain’s lifetime.

There are no emissions to net against, so the supply curve is flat rather than inflating or deflating. Shipping more Tier-1 features (paymasters, cross-chain messaging, intent bridging, oracle updates, randomness submissions, consumer-app activity, dApp deployments) raises gas usage, but gas usage feeds the base fee, which is destroyed rather than paid to the proposer. It does not by itself feed the burn.

This is the "ultrasound money" dynamic Ethereum got post-EIP-1559, and it is worth being exact about where Aere actually sits relative to it. Aere has not reached that crossover and is not near it. Reaching it requires validator coinbase revenue to be non-zero, which today it is not: genesis declares no blockreward, and the base fee is destroyed by EIP-1559 rather than paid to the proposer. There is a third fact that decides whether volume alone can close the gap, and it is worth stating precisely rather than hand-waving. Under EIP-1559 a validator's coinbase revenue comes from the priority fee, the tip, not from the base fee. Sampled directly on chain 2800, every recent transaction pays effectiveGasPrice exactly equal to baseFeePerGas, meaning a tip of zero. So while tips stay at zero, more dApps and more users do not move the crossover point, because 37.5% of zero is zero at any volume. Volume starts to matter only once transactions actually pay a priority fee, or once a block reward is introduced. The mechanism is live, immutable and correct; what it lacks is revenue to act on.

Where this combines with the rest of the session's work

Tier 1.3 Paymasters, every paymaster pays AERE gas → splits → burns. A high-volume paymaster like AereTokenPaymaster is a continuous deflation pump.
Tier 1.4 Cross-chain messaging, every Hyperlane message dispatch pays AERE gas → burns.
Tier 1.5 Across intents, every solver fill + every settlement message pays AERE gas → burns.
Tier 1.6 Pyth oracle updates, every updatePriceFeeds call pays AERE gas → burns.
Tier 1.7 drand submissions, every submitRound call pays AERE gas → burns.

Each shipped feature isn't just a capability, it's a stream feeding the deflation flywheel.

What's still pending

Foundation deploys the forwarder daemon to each of the 7 validator hosts. The Dockerfile + Python script ship with this release; deployment is ops work, ~30 min total.
Phase 2: move enforcement to consensus layer. Future Besu fork (or QBFT-extension proposal) where the chain itself splits coinbase per block. Removes the "validator could disable daemon" trust assumption. Until then, public accountability is the enforcement mechanism, and chain data is public.

What's next

Tier 1.9 ships batch-auction settlement on AereSwap for MEV-resistant trading. Tier 1.10 adds Fee Monetization / Sequencer Fee Sharing, developer revenue share, strong builder magnet. Tier 1.11 tunes sub-second block times.

AERE's economic engine now points the right direction: the destruction path is built, sealed and pointed the right way. What it still needs is transaction volume, because the burn is proportional to validator reward and validator reward is currently zero. The flywheel is assembled; it is not yet turning.

← Back to all posts