Farm Volatility

What Mill Finance is

Mill is a volatility-harvesting protocol on Robinhood Chain, and an openly derivative fork of Peapods Finance. You deposit a token and receive a milled token that is a claim on it. The claim is redeemable forever, and the amount it redeems for only ever goes up.

Why it exists

Yield that is not emissions. Most on-chain yield is a token being printed and handed to you. You are paid in the thing whose supply is increasing to pay you, which is dilution with extra steps. Mill pays holders by destroying supply: the fee is charged in milled tokens, most of it is burned, and burning raises what every remaining token redeems for. Nothing is printed. No one is diluted. The yield is denominated in the asset you already wanted to hold.

Exposure to a tokenised asset, plus the value of its own volatility. A tokenised stock on-chain does not trade in lockstep with the stock. It drifts, gets repriced, gets arbitraged back. Every one of those movements is trading activity that nobody currently captures. Wrap the tokenised stock in a Mill and that activity becomes revenue for the people holding it, while the exposure to the underlying stays intact.

The four pieces

A Mill is the vault. One Mill wraps exactly one underlying, permanently. It holds deposits, mints and redeems milled tokens, charges a fee on the way in and on the way out, and has no path by which anyone - including its owner - can move the deposits anywhere except back to a redeemer.

A milled token is what you get. mTKN for a Mill over TKN. It is a plain ERC-20, and it carries a redemption ratio that starts at 1.000000 and rises.

Volatility farming is what happens in the milled token's own market. mTKN trades against something other than TKN, so it can drift away from what it is worth. Closing that gap means wrapping or unwrapping, and that pays the fee. Nobody has to be right about direction - they only have to trade.

The flywheel is what connects them:

• Volatility moves the milled token away from its backing.
• Arbitrageurs close the gap by wrapping or unwrapping, which charges a fee.
• Most of the fee is burned, which lifts the redemption ratio for every holder.
• Part goes to liquidity providers, which deepens the pool.
• A deeper pool means a tighter band, which means the next gap gets closed sooner and more often.
• The rest builds the treasury, which funds protocol-owned liquidity - deepening the pool again.

Each turn makes the next turn cheaper to execute. That is the whole design.

The four invariants

• There is no path out of a Mill except redemption. No pause, no upgrade, no rescue, no owner function.
• The redemption ratio never falls, under any operation.
• Redemption can never be disabled. Deposits can be closed once, permanently.
• The burn floor is immutable and set at deployment. Governance can raise the holders' share and can never lower it below that floor.

Network

Not financial advice. Nothing here is an offer of anything, and nothing here is a promise of a return.

Milled assets

A milled token is an ordinary ERC-20. No transfer hooks, no rebasing, no balance that changes while it sits in your wallet. Anywhere an ERC-20 works, it works: a lending market, a vault, a multisig, a bridge, another AMM.

What makes it a claim is the redemption ratio - the vault's holdings divided by what it owes. It starts at 1.000000 and is monotonically non-decreasing: every burn pushes it up and there is no operation, admin or otherwise, that can push it down. Your balance never changes. What each unit of that balance is worth does.

Redemption is unconditional and permanent. New deposits can be closed once and for all by governance, for a delisted or deprecated underlying. Redemption cannot be closed by anyone, ever.

Wrapping and unwrapping

Wrap deposits TKN and mints mTKN priced at the current ratio, less the wrap fee. Unwrap burns mTKN and returns TKN at the current ratio, less the unwrap fee. Both settle the vault's outstanding fees before pricing, so nobody is ever quoted a stale ratio - and your own toll never partly refunds itself.

The fees are fixed at deployment and cannot be raised afterwards by anyone. Neither can exceed its hard ceiling: 1% to wrap, 3% to unwrap.

Milled tokens carry three more decimals than their underlying - 21 for an 18-decimal token. That is a deliberate virtual-share offset that makes the classic first-depositor inflation attack cost the attacker roughly a thousand times what they could steal.

Two ways to hold it

Passively. Hold mTKN and do nothing. The ratio rises under you as the mill turns. No staking, no locking, no claiming, no transaction of any kind.

As a liquidity provider. Supply mTKN and the quote token to the milled token's pool, then stake the position. Staked positions earn a share of every fee the Mill charges, on top of the trading fees the pool itself pays and the ratio growth on the milled half of the position.

Liquidity added through this app is always full range, the way a Uniswap v2 pool works. Only full-range positions can be staked: the vault weights positions by raw liquidity, and a narrow position carries far more liquidity per pound deposited, so allowing mixed ranges would let a one-tick position farm the whole reward.

A newly staked position waits five minutes before it can be activated and start earning. That window is what stops a bot minting liquidity in front of a large wrap and burning it the block afterwards - it has to hold real liquidity across a period it cannot predict, which is simply being a liquidity provider.

Why the gap exists

A milled token has two prices. There is what it is worth - the redemption ratio, arithmetic on what the vault holds. And there is what it trades at in its own pool, which is whatever the last trade left behind.

Those two numbers come apart constantly. Somebody sells size into a thin pool. The underlying moves and the milled token's pool has not caught up. A large holder rebalances. None of it requires anyone to be wrong - it is the ordinary behaviour of a market against a reference price that moves independently.

Trading below redemption value

The milled token is cheap. Anyone can buy it and redeem it for more than they paid.

1. Buy mTKN in the pool, below its redemption value. Buying pushes the pool price up, toward backing.
2. Unwrap it. The vault returns TKN at the full ratio, less the unwrap fee.
3. Keep the difference. The trade is profitable whenever the discount exceeds the round trip: pool fee, plus the unwrap toll, plus slippage.

The pool's hook charges nothing on the direction that buys the milled token. A discount is the state that makes a backed wrapper look broken, and it is repaired by exactly this trade. Taxing it would be taxing the repair crew.

Trading above redemption value

The milled token is expensive. Anyone can mint it for less than the market will pay.

1. Buy TKN, or use what you hold.
2. Wrap it into mTKN at the ratio, less the wrap fee.
3. Sell the mTKN into the pool, above backing. Selling pushes the pool price down, toward backing.

This direction pays the pool fee, the wrap toll and the hook's directional fee, which is charged only on the side that sells the milled token. A premium is a benign state and the protocol is happy to charge for the privilege of unwinding it.

The band

Between those two thresholds nothing happens, because nothing is profitable. That gap is the no-arbitrage band, and its width is the sum of the round-trip costs on each side. Inside the band the price simply is where it is.

The band is not a defect - it is the price of the mechanism. What matters is that it is narrow, and it narrows as liquidity deepens, which is precisely what the LP share of the fee is for.

What each crossing produces

Every time the band is crossed, someone wraps or unwraps. That is not a side effect - it is the only way to close either gap. So each crossing yields, in the same transaction:

• Fee revenue, charged in milled tokens.
• Ratio growth, because most of that fee is burned, and burning raises what every remaining token redeems for.
• Price stability, because the arbitrageur's trade is what moves the pool back toward backing.

The three are the same event seen from three angles. This is why volatility is the input: a quiet week produces no crossings and no yield, and a violent one produces a great many of both.

Milled tokenised stocks

A tokenised stock is an on-chain instrument that tracks a real one. It is not the same thing as the stock, and it does not trade in lockstep with it. It has its own order book, its own liquidity, its own hours, and its own reasons to move.

Wrapping one in a Mill keeps the exposure and adds a second source of return. Hold mSPY and your backing is SPY: if the tokenised stock doubles, so does what your position is worth. The ratio grows in units of the stock, not in dollars - a Mill does not protect you from the underlying falling, and it is not trying to.

Where the activity comes from

A tokenised stock has two clocks, and a milled one earns from both.

On-chain. The tokenised stock trades continuously against crypto assets, including outside market hours, driven by on-chain flow that has nothing to do with the equity. Every move repositions the milled token's pool relative to its backing, and closing that gap runs through the mill.

Off-chain. The real equity moves while the chain is not watching. When the tokenised version has drifted from the stock it represents, the gap does not close by itself - it is closed by a licensed market maker, who is authorised to create and redeem against the real shares and does so when the discrepancy is worth acting on.

Two moments, both of which turn the mill

That produces a repeating shape, and both halves of it generate flow:

1. The mispricing. The tokenised stock drifts from the equity it tracks. Traders act on the discrepancy, which moves the tokenised price - and moves the milled token's pool away from its backing.
2. The re-peg. The licensed market maker steps in and brings the tokenised price back to the real one. That is a second move, in the opposite direction, and the milled token's pool has to be arbitraged back again.

A wrapper on an asset that only ever moved once would earn once. A wrapper on an asset that is systematically pushed away from its reference and then systematically pulled back to it earns on both legs. The mispricing and its correction are the same round trip, and the mill charges for the whole of it.

This is why tokenised equities are the natural fit. The re-peg is not speculation about what a market maker might do - it is the mechanism by which a tokenised stock tracks its underlying at all, and it is what makes the asset worth holding in the first place.

$MILL

MILL is the protocol's own token, launched on Pons v2 and paired against wETH.

The launch fee

Trades in the MILL pool carry an additional 0.5% fee that goes to the protocol. It is charged in the pool, not on transfer - MILL moves as a plain ERC-20 in a wallet, and the fee applies only to trading against that pool.

That fee builds the treasury, which has two jobs:

• Running the protocol. Audits, infrastructure, and the work of building the next mills.
• Protocol-owned liquidity. Liquidity the protocol holds itself rather than renting from mercenary capital. Rented liquidity leaves the moment the incentive stops; owned liquidity does not, and depth is what keeps the arbitrage band tight enough for the whole mechanism to work.

The second is the one that compounds. Every mill that goes live needs a pool deep enough to arbitrage against, and a treasury that can seed one is the difference between launching a mill and launching a mill that turns.

What holding it is

MILL is the claim on the protocol's revenue and on its governance. It is not a claim on any Mill's deposits - nothing is, except the milled tokens themselves. Governance can move the fee split above the immutable burn floor, change where the protocol's share is sent, and close new deposits on a mill. It can never reach the assets inside one.

Where protocol revenue comes from

Three streams, all of them charged on activity rather than printed.

1. Trading fees on milled asset pools

Every milled token pool carries a Uniswap v4 hook that takes a fee on the direction which sells the milled token, and nothing at all on the direction that buys it. That fee goes to the protocol.

This is not the pool's own trading fee. The pool fee belongs to the liquidity providers and Mill takes none of it. The hook's fee sits on top, is capped at 5% by a constant no one can raise, and every change to it - in either direction - serves a five-minute delay, so the rate you can read on-chain is the rate a swap will actually be charged.

It is directional for a reason. Buying a discounted milled token is what repairs the peg; selling into a premium is what unwinds a benign state. Charging the second and not the first means the protocol earns from the trade it can afford to tax.

2. The protocol's share of the Mill fee

Every wrap and unwrap charges a small fee in milled tokens, split three ways:

• Burned - destroyed, which raises the redemption ratio for every holder. This is the largest share, and its floor is fixed at deployment and can never be lowered.
• Liquidity providers - paid to whoever has staked a position in the milled token's pool.
• Protocol - the remainder.

Governance can move revenue between the last two freely and can always give more to holders. It can never take holders below the floor.

3. The $MILL launch fee

The 0.5% on trades in the MILL pool, described under $MILL.

What the protocol does not earn

Nothing from deposits sitting still. There is no management fee, no performance fee, no yield on idle assets, and no rehypothecation. A Mill that nobody trades through earns the protocol nothing at all - which is the correct incentive, and the reason the roadmap is about liquidity depth rather than about total value locked.