Moneywith memory.

Price attracts attention. Attention attracts capital. Capital deepens liquidity, and liquidity attracts more of all three. Every serious attempt to engineer this out of crypto has produced something nobody wanted to hold.

Ducat does not try. Reflexivity is the most reliable capital formation mechanism this market has ever produced, and a protocol that treats it as a defect is refusing free money on principle.

The defect is narrower than that. A token can travel from a dollar to twenty dollars and back to a dollar while leaving nothing behind. The run happened. The protocol has the same balance sheet it started with, and everyone who arrived late paid for the education.

The problem is not reflexivity. It is reflexivity without memory.

Ducat is built to remember. Every run it survives leaves it holding more than it held before, and the record of that is not a chart or a narrative. It is a balance sheet, published, marked down and readable from the chain.

A banking experiment, run in the open

Strip away the token and what is left is a question worth asking plainly: what happens when a central bank publishes its whole balance sheet, writes its rules into code it cannot quietly amend, and gives itself exactly one mandate, to get heavier over time?

That is the experiment. Not a coin, not a chart, not a community. A monetary authority with no boardroom, tested the only way a monetary authority can be tested, by having its balance sheet read by anyone who wants to and its rules run whether or not the moment is convenient.

Ducat did not arrive at this from nothing. It is downstream of two prior attempts to answer the same question, each of which got something importantly right and something importantly wrong. Section 24 traces that lineage in full. What matters here is the posture it leaves Ducat with: a treasury that is a balance sheet rather than a story, and a currency whose rules do not bend to whoever is in the room when things get hard.

The shared recovery regime is a planned extension. The coordinated bond gate, hook-published regime and inverse bonds described below still require final implementation and activation rules.

Most currencies are asked to hold still, and spend enormous effort failing to. Ducat is built to move, in both directions, on a schedule nobody sets by hand. What decides which direction it is moving in at any moment is a single published ratio: the stablecoin reserve held against every ducat outstanding, compared to the reference. Above 1.20 times cover, the system expands. Below it, the system contracts. Nothing else votes on which state it is in, and nothing needs to.

Call it a business cycle compressed into a set of contracts and stripped of the discretion a human central bank would normally exercise inside it. The shape is not new. Good conditions build a cushion. The cushion is what the system draws on when conditions turn. What is different here is that the line between the two states is a number anyone can read on chain, rather than a judgment call made by a committee after the fact.

Expansion above 1.20× cover

Bonds sell into a rising premium and bank the proceeds permanently. Free issuance mints against a surplus that has already arrived. The ratchet, reading a sustained low rather than a single print, becomes eligible to move the reference up a rung. Every one of these adds to the reserve, or retires a claim against it, and none of them can run in the other direction.

Contraction below 1.20× cover

Bonds and free issuance both stop at the same line, together. Inverse bonds open a standing buyback that burns ducat out of the market. The desk turns overweight equity back into USDG. Redemption keeps paying, at the lower of the reference or the reserve, so the exit never becomes the thing that empties the treasury. Nothing here waits for a vote.

The two states are not competing designs bolted together. They are one design, read from two sides of the same threshold. A protocol that only knew how to expand would be a bull market with a smart contract attached. A protocol that only knew how to contract would never grow a treasury worth defending in the first place. Ducat is built to do both, and to do them without anyone having to decide, in the moment, which one applies.

The system does not choose a season. It reads the reserve and finds out which one it is already in.

Everything from here forward belongs to one side of this line or the other. Sections 6 and 7 are expansion in mechanical detail; section 16 and the redemption right in section 18 are what contraction looks like when the line is crossed. Section 23 draws the whole cycle out end to end, in both directions.

Market can rise and fall freely. Backing fluctuates as treasury assets appreciate, depreciate and produce income. Reference moves deliberately, and in one direction. Only the ratchet contract can write it, and the contract rejects any value that is not strictly higher.

A sustained reserve can justify a higher reference. A speculative spike cannot. The asymmetry runs the other way too, and this is the part that matters: if the reference has reached $1.2155 and the reserve later falls to $1.10 per ducat, the protocol does not declare $1.10 the new reference and carry on. Issuance has already stopped, redemption pays the lower of the two, and every safeguard that reads the reserve is acting on it.

A ratchet that moves both ways is not a ratchet. It is a chart.

This is what makes the reference worth measuring anything against. Price can go backward. Backing can go backward. The monetary objective does not follow them down, which means every level it has reached is a level the system has committed to defending rather than a high water mark it once printed.

Bonds are the converter between reflexivity and balance sheet growth. They are also the most direct way to thin a balance sheet out, and the difference is where the line is drawn.

If backing sits at $3.00 per ducat and the protocol issues new units for $1.50 of assets each, it has grown the treasury and diluted every existing claim on it. The headline improves. The cushion per ducat gets thinner.

Ducat allows that, up to a line. Bond pricing reads the time-weighted market price and the reference. It does not read backing. A bond can be sold below backing. It can never be sold below the reference, and after every bond the reserve must still cover the reference on the whole supply, checked on the state after settlement rather than before it.

A bond paid in stocks carries one more rule. The reference cost of the ducats it issues comes off the surplus that issuance reads, so the same room is never spent twice.

The cushion is allowed to flex. The promise is not.

Accumulating assets is the first stage and the least interesting one. Reserves that sit still depend entirely on the next wave of demand to grow. Ducat is built so the balance sheet keeps earning after the attention leaves.

The treasury holds stablecoins and tokenised equities the protocol owns outright, each marked down by its own haircut before it counts toward backing. At least 35% of the treasury must be instantly payable stablecoins, and redemption pays from that leg alone, so the exit stays payable while equity markets are shut.

Haircuts price what can go wrong with a mark, not the asset class. Every position takes 1% while its feed is fresh. A stablecoin whose feed goes stale takes 5%, and a stock whose market is closed takes 10%. A non-stable position that grows past 40% of its class carries a surcharge of up to 25% on top. Stocks do not need a deep class haircut, because they never back the promise.

Figures illustrative. Every position is published, and the haircut is applied before the asset is allowed to count.

How the stock book is sized

The protocol never spends on stocks. The equity book is an allocation of excess reserve, and it arrives by bond. Its target is a share of the stablecoin surplus above 1.30× cover, rising to half of the excess at 2.00×, split AAPL 34, NVDA 33 and TSLA 33. Bonds bring stocks in, and their discount widens with the gap to target.

The way out needs nobody's permission either. When a stock runs over its weight, anyone can buy the excess from the desk at the oracle price plus 1%, paying in USDG, which is kept, or in ducat, which is burned.

Paths that move value use strict price reads that revert on a stale or paused feed, so a stock cannot be bonded or bought from the desk at a price nobody is publishing. Both close while its market is shut. Valuation and redemption carry on at the stale haircut.

What the balance sheet earns

Stablecoins above the liquid floor can be placed in approved vaults, and the vault shares still count toward the reserve. That is the implemented income path, alongside the buy-side toll on the pool and the desk's margin; this paper does not establish live deployment. Equity distributions are not built. None of it is paid out.

Further out, gauge weight can direct voting positions across aligned markets and earn emissions and incentives there. A gauge controller and a sync to an external voter already exist. The voter is unset at deploy. That layer is the roadmap rather than the running system, and the paper marks it as such.

The implemented core is the part that matters most: the protocol owns its assets outright, marks them down honestly, promises only against its stablecoins and publishes the result.

Protocol-owned liquidity, coming to a second pool

The reviewed launch configuration assigns no protocol-owned position in the canonical DUCAT/USDG pool. A second pool, DUCAT against tokenised equity, is being stood up alongside it, and this one the protocol seeds and keeps a position in.

Owning that liquidity is what makes the pair worth having. A market with no protocol depth behind it is thin exactly when it is most useful, and a thin pool is a bad venue for the cross-pool arbitrage the second pool exists to enable, covered in section 19. Protocol-owned liquidity fixes the depth problem at its root instead of subsidising it week to week.

The position is not free money sitting in a vault. It is a liquidity position, redeemable only by withdrawing it, so it carries its own tier and its own haircut and does not count toward the 35% liquid reserve floor. Where its funding is drawn from, and the exact haircut it carries, are being finalised alongside the levy in section 19.

Issuance weight and gauge weight are separate fields in the contract, and escrow keeps a ledger for each. The founder rank is where the reason becomes visible: ten times the economics, three times the say.

One weight answers how much of permitted expansion a pass participates in. The other answers how much influence it has over where the protocol's liquidity power is directed. Fusing them is the standard design, and the standard result is that whoever arrived first controls everything forever.

Keeping them apart lets the rare pass carry the economics of a rare rank without carrying proportional control over the treasury. Rarity stays desirable. The system stays governable.

Gauge voting is implemented over a governance whitelist of gauges. The position it will direct is not: the external voter it syncs to is unset at deploy.

A pass is bought into the reserve. It can also be destroyed for a slice of the equity sleeve, pro rata by rank, which gives the ownership layer an exit that does not require anyone to buy the pass from you.

The slice is weighted by issuance weight, because that is the weight that describes the economics. A founder burning against ten units of weight receives ten times what a member receives against one, at the same moment, from the same pool. It is the ordering the expansion split already uses, applied to the way out rather than the way in.

What it pays out of is deliberately narrow. The claim reaches the tokenised equity positions and nothing else. It pays in kind, the shares themselves, less a flat 10% that stays in the treasury. The in-kind payout does not require an equity price quote, so market closure alone need not prevent it, and the stablecoin reserve is out of reach by construction rather than by a test.

A pass has no claim on the promise. It has a bounded claim on the equity book.

That distinction is the whole safety argument and it should be read carefully. Paying shares out against a burned pass removes treasury value without removing a single monetary claim, so backing per ducat falls by construction. There is no arrangement of this mechanism that makes that untrue, and the paper is not going to pretend otherwise. What it cannot lower is the reserve per ducat, because it never touches a stablecoin.

So the exit needs no floor. There is nothing a solvency test would be protecting. It is metered by a daily allowance per asset and remains subject to treasury pause state, asset availability and successful token transfers. An empty equity book supplies no payout; a holder should not burn a pass for a zero-value exit.

1. Pro rata by rankWeighted by issuance weight against the weight still outstanding, paid in kind from the equity book less the exit fee.
2. Metered, not gatedA daily allowance per asset and one governed fee. No reserve-cover gate is required, but pause controls and exhausted allowances can still prevent execution.
3. A worked exampleA Founder burns against 10 units of weight, out of 800 outstanding, when the equity book holds $2,000,000. That is 1.25% of the book, $25,000 of shares before the fee, $22,500 after the 10% held back. An alternative Member burn against that same pre-burn snapshot, at 1 unit of weight, receives a tenth: $2,250. These are separate examples, not sequential payouts; the first burn changes both balances and outstanding weight. Neither example touches a stablecoin.
4. The pass does not come backA burned slot is never minted again. The count falls permanently and total issuance weight falls with it.

The same terms in every weather

Closing the exit in a downturn would be the easy version, and here it would protect nothing. The exit cannot touch what backs the promise, so its terms do not change through a cycle: one governed fee and per-asset daily allowances, without an additional reserve-cover gate. Operational pauses and transfer restrictions still apply. An ownership exit that vanishes the moment holders want it was never an exit.

The step worth sitting with is the last one. Expansion is split across passes by weight, so retiring weight raises every surviving pass's share of every expansion the protocol ever makes again. The holder who leaves takes equity. The holders who stay take a permanently larger claim on the growth of the monetary base, and they do not have to do anything to receive it.

It is the same shape as the exit one layer up. Redemption pays the lower of the reference or the reserve per ducat, less a fee, and leaves the remaining holders no worse capitalised than they were. A sleeve burn pays shares less a fee and leaves the remaining passes holding more of the future. Both exits are real, both are capped, and both are priced so that using one is a transfer to the people who did not.

Between them the ownership layer has a floor that does not depend on anyone wanting a pass. Passes are bought into the reserve, which makes the currency heavier. They are burned for equity, which makes the survivors' claim larger. The drawn count only ever moves down.

The exit fee is one governed number, capped at 50%, and the daily allowance per asset is another. Any revealed, unlocked pass is eligible. Both settings sit in the public queue before they take effect.

The defence against a downturn is not a stock of the protocol's own token. No contract holds ducat against itself. The defence is stablecoins, built in the good months by what issuance is not allowed to spend.

Expansion issues 70% of each rise in the stablecoin surplus and leaves the rest where it is. Nothing is issued at all below 1.20× cover. Every buy on the pool pays 0.5% into the reserve, and every sell burns 1% of the ducat sold. Yield on stablecoins above the liquid floor accrues to the reserve and is never paid out.

A strong market therefore finishes with more stablecoins behind every ducat than it started with, and a weak one draws on a cushion that was built for it. Every print goes to the protected vault and to locked passes, and a period in which nobody is locked sends the pass leg to the treasury.

The mechanism is countercyclical by construction rather than by anyone's discretion at the moment it is needed, which is the only time discretion has ever failed.

The shared regime flag belongs to the revised design. Price observation and issuance checks already exist; the shared flag is a planned extension.

Policy is not here to hold a line. It reads the market and the reserve, then allows exactly as much issuance as the reserve has already paid for.

The hook under the ducat's own Uniswap v4 pool on Robinhood Chain does three things on every swap, not one: it records the price observation, it takes the toll, and it checks the reserve against the 1.20× line and flags which regime is currently live, expansion or contraction. Every other contract that cares which regime is live, bonds, free issuance, the levy split described in section 19, reads that flag rather than recomputing the check itself, so the two states described in section 3 are read from a single place on chain rather than assembled independently by whichever contract happens to need the answer. The issuance decision itself is a plain contract that reads a 30 minute time-weighted price from the same hook and asks the market one question: is it at or above the reference, and within 10% of its seven day anchor, on a pool with at least $8,000 of vouched depth?

The answer is yes or no. How much follows from the stablecoin surplus and the standing rate. Composition limits sit on bonds, not on the gate. Volatility and bond demand are not inputs.

There are no bands. A ladder of regimes saturates at every price a run produces and tells policy nothing the level test does not.

A one block spike and three weeks of sustained bid are not the same event, which is why the gate reads a thirty minute average and holds it against a seven day anchor rather than trusting spot. The size of a premium is not an input at all. A $3.75 market against $1.60 of stablecoins and a $3.75 market against $1.05 are entirely different systems wearing the same price, and it is the reserve, not the premium, that decides how much either may issue.

Unreadable fails open, dark fails closed

A pool that has never been readable, because it is too young or thinner than $8,000 of vouched depth, does not gate issuance. A pool that was readable inside the anchor window and then stopped closes issuance and bonds until it returns. A thin pool is not a cheap off switch, and a drained pool is not a cheap on switch.

This is not discretionary central banking. Nothing here is a judgement call made in the moment. Policy cannot declare backing, which is computed from oracle reads. It makes no treasury call at all. It cannot mint without limit, because every mint passes the token's own cap and a solvency check after it settles. It cannot ratchet the reference on a single market move, because the ratchet reads a trailing window. Its authority is real and its perimeter is fixed.

Policy is powerful because its powers are bounded.

The existing pool toll and proposed unified levy are separate versions. The second pool, levy rates, decay window, funding split and liquidity-position haircut remain pending. The proposed 5% ceiling is not the current hook configuration.

The ducat trades on one canonical pool, ducat against USDG, and the hook under it takes a toll in both directions. 1% of every ducat sold is burned. 0.5% of every USDG spent on the ducat is deposited to the treasury. Both rates are governed through the timelock and capped at 3%.

Which happens is fixed by direction. Sold ducat is burned and spent stablecoins are held. There is no treasury decision to make and no discretionary pot in between.

1. Paid by everyoneThe one percent is a tax on selling ducat, paid by every seller, not a fee on one kind of trade. Nobody is excluded, and everybody contributes.
2. Held or burned, by directionBuys pay stablecoins into the reserve. Sells destroy part of what is sold. Nothing buys ducat on the pool to burn it.
3. Both routes land in the same placeHeld, the reserve grows against unchanged supply. Burned, supply falls against an unchanged reserve. Stablecoins per ducat rise either way.

Nothing in this line is paid out. It goes back to the currency and to nobody else. Pass economics come from permitted expansion, where they are defined, weighted and junior by design. Keeping the two apart means the toll never has to be argued about in a drawdown, because there is no claim on it to suspend.

The toll also costs something, and that should be stated rather than discovered. A trader who buys and sells back pays 1.5% on the round trip. That is the price of a pool that feeds the reserve on every trade.

The deeper corridor

There is a corridor underneath the pool that does touch the balance sheet. When the market trades below the redemption price, buying on the pool and redeeming at the lower of the reference or the reserve, less the fee, is profitable, and stays profitable until the gap closes. That is the corridor that makes the floor real rather than stated, and anyone with ducat and a pool to buy it on can work it.

The two rates on the canonical pool are governed settings and sit in the public queue before they take effect. Whether proceeds are held or burned is not a setting.

A second pool, and the arb it brings back

One pool cannot disagree with itself. A second is being stood up, ducat against tokenised equity, and the protocol seeds it and keeps a liquidity position there rather than leaving the depth to whoever shows up, as section 9 describes.

Two pools pricing the same asset will not agree for long, and that disagreement is worth recovering rather than leaving on the table. The corridor this reopens is the one the original design called the spread between the pools: buy the ducat where a pool implies it is cheap, sell where another implies it is rich, close the cycle flat in ducat and ahead on the difference. It needs both pools to be real markets with real depth, which is exactly what owning the second pool's liquidity is for.

What is captured this way does not sit in a discretionary pot, and it does not sit outside the fee structure either. It is priced and collected through the same levy described below, on whichever pool the dislocation is closed against, and it is distributed exactly the way every other line under that levy is distributed: part to the reserve, part to funding the protocol, on the schedule set out next. A trade that closes a dislocation is still a trade, and it pays the same toll a directional one would.

The levy: one number, capped, and decaying

A single levy is replacing the pieces above with one published rate, enforced the same way it is today: the hook. It deploys on the new equity pool the same way it deployed on the canonical one, wired in from that pool's creation, so both venues run the identical contract rather than two different fee mechanisms. Two things about the rate are fixed now, even though the number itself is not.

First, the ceiling. Combined, across whatever the levy eventually is and whatever of the original toll survives inside it, the total a trade can be charged will never exceed 5%. That is a hard cap this document is committing to before the rate underneath it is finalised, not a target the rate happens to land near.

Second, the shape. The levy opens at the top of that range and decays down to its resting rate over a period governance will set and publish before launch, rather than sitting flat forever. Early trading carries the heavier load, when the reserve is thinnest and the protocol's own build and operating costs are least offset by anything else. As the reserve deepens, the rate eases down, and the split underneath it moves too: more of the levy goes to protocol management at the start of that window, and more of it goes straight to the reserve as the window closes, ending, like the pool toll it replaces, almost entirely as a reserve inflow.

What the rate is, on day one and at rest, and how long the decay window runs, are being finalised. This document will carry the numbers once they are set, in this section and in the params table in section 21.

None of this touches the invariants in section 20. The levy funds itself out of trading activity, never out of the reserve directly, and it is not a claim anyone junior can be paid ahead of the currency itself.

Every constraint described above is a published parameter with a value, not a principle with a tone of voice. This register combines the reviewed launch settings with the revised design. Shared recovery controls, the second pool, unified levy and $10.00 to $0.50 ticket clock remain proposals pending final configuration; they are not live readings.

1. 1$1.00
   Opening referenceWhere every ducat's redemption is measured from on day one. It can only go up, and only through the ratchet.
2. 2$3.00
   GenesisA 24 hour fixed-price sale in USDG, capped at $30,000 and $1,000 per wallet. Mints the first ducat at settlement, banks 70% and seeds the pool with the rest. Settles only at 1.20× cover or better, and a raise too small to seed a readable pool refunds instead.
3. 31,000,000 a day
   Epoch mint capEnforced by the token over a sliding 24 hour window. Sized to admit the genesis settlement and little more. Raising it is timelocked.
4. 421 epochs
   BootstrapFor the first 21 eight hour epochs after the timelock delay, issuance has no rate cap. Only the surplus rule and the cover floor bind. The end is immutable.
5. 510.5% weekly-equivalent
   Expansion capA standing issuance rate applied per epoch, for bonds and free issuance together once the bootstrap ends. It is not a separate rolling-week hard cap. Issuance spends 70% of each rise in the stablecoin surplus and nothing below 1.20× cover.
6. 61.20× on stablecoins
   Recovery thresholdBelow this, bonds and free issuance both stop, and the inverse bond and desk-to-USDG levers described in section 16 become active.
7. 71.00× on stablecoins
   Bond floorThe reserve must cover the reference on all supply after every bond. The margin scales with the reference automatically.
8. 8TBD
   Inverse bond pricingThe rate at which the treasury buys back and burns ducat below 1.20× cover. Sits alongside the bond floor as its mirror; not yet set.
9. 965%
   Bond capacity ceilingOf the standing allowance per epoch, opened market by market by governance.
10. 10min(reference, reserve) less 1%
    RedemptionPaid in USDG from the liquid leg, up to 3% of the liquid reserve a day. Every redemption must leave the reserve per remaining ducat no lower. The fee is what makes every exit accretive to the holders who stay.
11. 115% step, once a day
    RatchetNeeds a five day low of stablecoins per ducat at 1.25× the new reference, so a raise to $1.05 requires $1.3125. The ratio is immutable.
12. 12E × P / (S + P)
    Expansion splitThe protected vault's share of every print. Locked passes take the rest. No liquidity slice and no bond slice. The treasury receives ducat only in a period when nobody was locked.
13. 1335% of treasury
    Liquid reserve floorInstantly payable stablecoins only, so redemption never waits on an exchange opening.
14. 141%
    HaircutOn every asset with a fresh feed. 5% for a stale stablecoin feed and 10% for a stock whose market is closed, plus up to 25% on a non-stable position over 40% of its class.
15. 151% sell, 0.5% buy
    Pool tollDucat sold is burned. USDG spent is held. Each rate is capped at 3%. The levy below is expected to fold this in.
16. 16One, becoming two
    PoolsThe canonical DUCAT/USDG pool, plus a DUCAT/equity pool now being stood up with a protocol-owned liquidity position, sized and funded as described in section 9.
17. 175% ceiling
    LevyA single trading fee across both pools, split between the reserve and protocol management. Never exceeds 5% combined. Opens at the top of that range and decays to a lower resting rate over a window still being set; the split shifts toward the reserve over the same window. The exact rate and decay period are not yet published.
18. 18Same as the levy
    Arbitrage captureSpread closed between the two pools is priced and collected through the levy above, on whichever pool it is closed against, and distributed on the same reserve and management split. No separate pot, no separate rate.
19. 195 days
    Bond vestSet per market, and the first market vests over five. Supply reaches circulation gradually. The treasury receives the assets in the same transaction.
20. 2040 / 80 / 880
    PassesFounder, Charter and Member, drawn by verified randomness, weighted 10/3, 4/3 and 1/1. Recovery 3/1, at zero supply until opened.
21. 21$500
    Pass eligibilityBonded over the last three completed eight hour periods, or a governance whitelist.
22. 22$10.00 to $0.50
    Ticket clockPer eight hour period, one ticket per account, everyone settling at the clearing price. Thirteen tickets a period, halving weekly to a floor of four. Paid in USDG at launch.
23. 2310%
    Pass exit feeKept by the treasury on the shares handed over. Governed, at most 50%, with a daily allowance per asset.
24. 2490% of supply
    Protected vault capThree day exit cooldown. Fees of 3%, 2% and 1% apply before one, three and five days respectively, then zero. Collected fees stay in the vault.
25. 25Four burns
    Supply retirementThe sell toll, ducat paid to the desk, redemption, and ducat-paid pass periods, which are off at launch. Each is visible on chain as it happens. Nothing sums them into a published figure.

Governed settings sit in the public queue at least two days before they take effect. A shorter list is not reachable by governance at all: lowering the reference, the ratchet's funding ratio, the end of the bootstrap and the genesis price.

Every part of the design resolves into one sequence, and the sequence does not require the market to keep going up. It runs the other way just as cleanly when the market does not cooperate, which is the whole point of building it as a loop rather than a plan.

Neither half needs the other's permission to run, and neither is a failure mode of the first. A currency that only expanded would be a bull market wearing a balance sheet. This one contracts on the same terms it expands: mechanically, publicly, and without anyone having to be right about where the cycle is headed next.

The objective is not perpetual price appreciation. No design delivers that and every design that claims to is lying. The objective is perpetual improvement in the machine underneath the price.

Crypto forgets quickly. Narratives rotate, premiums evaporate, liquidity migrates, and assets that traded at extraordinary valuations return to earth about as fast as they left it.

Ducat does not try to prevent any of that. It tries to remember it.

When enthusiasm arrives, the protocol converts part of it into assets. When enthusiasm leaves, the assets stay and keep earning. Backing records the result, marked down and published. And when the reserve has sustainably proved the system deserves more, the reference steps up and stays there.

The goal was never to hold the ducat at $1.00.

The goal is to make $1.00 the first chapter.