21 million is sound. The settlement layer got debased.
Bitcoin has two scarce resources. Everyone talks about one. No one talks about the other.
The first is 21 million coins. Hard cap. No inflation. The supply schedule is enforced by code. No central bank can print more. This is the monetary policy. This is the scarcity everyone knows.
The second is block space. This is the scarcity no one talks about. Every block is a settlement. Once a transaction is in a block, it is done. Irreversible. Final. The block is a one-way ratchet: once written, it can never be rewritten. This is what makes Bitcoin a settlement network, not just a payment network.
In finance, finality is everything. A trade is not settled until it is final. A payment is not done until it cannot be reversed. Finality is the thing that makes the system trustworthy. In Bitcoin, finality comes from block space. The scarcity of block space is what makes settlement scarce. The scarcity of settlement is what makes the system secure.
Monetary debasement is simple: more units of money, same goods. Each unit buys less. The supply of money inflates. The purchasing power of each coin falls. This is what central banks do. This is what Bitcoin was built to prevent.
But debasement is not limited to coins. Any scarce resource can be debased. The mechanism is always the same: inflate the supply, reduce the cost, push the difference to someone else.
Block space is a scarce resource. It can be debased. And it was.
Before SegWit, block space was simple. 1 MB cap. Every byte costs the same fee. Block space is tight. Settlement is expensive because it should be. Finality is scarce because it is.
SegWit changed the rules. The block went from 1 MB to 4 MB. Witness data — signatures, scripts, the cryptographic proof that a spend is valid — got a 75% discount. Same bytes. Same disk. Same bandwidth. Same cost to run a node. But the block carries 4× more data at 1/4 the price.
If someone inflated the Bitcoin supply by 4× and discounted new coins by 75%, you would call it monetary debasement. That is what SegWit did to block space. The settlement layer got bigger, cheaper, and the cost got pushed to node runners.
In monetary theory, the Cantillon effect describes who benefits from inflation. The first recipients of new money get it at full value. By the time it circulates, the purchasing power has fallen. The early receivers win. The late bearers lose.
SegWit created a Cantillon effect for block space. The first users of witness discount — exchanges, custodians, early adopters — got cheap settlement. They stored data at 1/4 the cost. The bill came later: node runners now store 4× more data for the same hardware. The discount was consumed early. The storage cost is forever.
The fee market prices inclusion. It does not price persistence. A writer pays a one-time fee to get into the block. The network carries the bytes forever. The cost of permanence is externalized to every node operator on Earth, in perpetuity. The writer pays for a quarter of what the network stores.
SegWit inflated the block. Taproot hid what is inside it.
Pre-Taproot, every UTXO shows its spending conditions on-chain. The ledger is complete. What you see is what exists. If you hold a Bitcoin, you can read the chain and know exactly what rules govern your coin. Who can spend it. Under what conditions. The full contract is visible.
Taproot changes this. You can build a tree of possible spending conditions off-chain and only reveal the one you use. The tree is real work. The hashes are real energy. The branches represent real alternative futures. But the chain does not record them. When you take the key path, the other branches vanish from the record.
This is off-balance-sheet. The chain carries liabilities it does not show. The ledger has a shadow. If you are a monetary theorist, you know what happens when ledgers have hidden liabilities. The 2008 financial crisis was built on them. Shadow banking. Off-balance-sheet vehicles. Obligations that existed but did not appear on the books.
Taproot introduced the same structure to Bitcoin. Not at the level of a bank. At the level of the protocol itself.
Here is the part that connects to physics, but it is really about money.
Every byte the network stores has a cost. Not a theoretical cost. A real cost. Disk. Bandwidth. Validation CPU. RAM. Every full node on Earth must accept, store, and serve every byte in every block, forever. This is not optional. This is what it means to run a node: you carry the chain forward.
When you write data to the chain, you are asking every node operator on Earth to store that data forever. The cost is not zero. The cost is paid by someone. In Bitcoin, the fee market is supposed to price this: you pay a fee, the miner includes your transaction, the network stores the bytes. The fee should reflect the cost of persistence.
It does not. The fee market prices inclusion — getting into the block. It does not price persistence — being stored forever. A transaction that pays 1 sat/vB costs the same to the network as a transaction that pays 100 sat/vB, once both are in the block. The network stores both forever. The cost is the same. The price is not.
The witness discount makes this worse. A witness byte costs 1/4 the fee of a non-witness byte. But it costs the same to store. The gap between what the writer pays and what the network costs is the externality. It is paid by node runners. It is paid forever. It is the hidden tax of block space debasement.
Try this thought experiment.
Case 1: Send a transaction. Cost = whatever the fee market says. A few sats. Maybe a few hundred. The fee market prices inclusion. This is what Bitcoin is designed for — circulation, transfer, exchange. 21 million circulate freely. This is monetary scarcity.
Case 2: Change block 210,000. Try to rewrite that block. The coinbase. The hash. The timestamp. Anything. You would need to re-mine every block after it. You would need more energy than the entire network has consumed since that block was written. You would need to double-spend history itself. The fee market cannot price this. The cost is not set by supply and demand. It is set by physics.
One cost is economic. The other is thermodynamic. One is a fee. The other is the energy to rewrite the universe.
This is what finality means. Block space is not just scarce. It is irreversible. Once written, it is part of the permanent record. This is what makes Bitcoin a settlement network. This is what makes it trustworthy. This is the scarcity that SegWit debased.
The pattern is familiar to anyone who studies monetary history.
A scarce resource exists. It is valuable because it is scarce. Someone finds a way to inflate the supply. The inflation is sold as an improvement. It is an improvement — for the early receivers. The cost is pushed to the late bearers. The late bearers are the node runners, the archivists, the people who carry the chain forward.
SegWit inflated block space. Taproot hid what is inside it. The settlement layer got bigger, cheaper, and more opaque. The monetary policy is still sound. The settlement policy got debased.
This is not a metaphor. It is the same mechanism that debases fiat currency, applied to a different scarce resource. The only difference is that fiat debasement is done by central banks. Bitcoin's settlement debasement was done by protocol upgrade. The effect is the same: the people who carry the cost are not the people who chose the inflation.
Sound money requires two things: a scarce supply and a trustworthy ledger. Bitcoin has both. The supply is capped at 21 million. The ledger is enforced by proof-of-work. No one can inflate the supply. No one can rewrite history.
But the ledger has a cost. The cost of storing, validating, and serving the chain is paid by node runners. The fee market is supposed to compensate them. It does not. The fee market prices inclusion, not persistence. The cost of permanence is externalized.
Sound money also requires a complete audit trail. Every obligation must be visible. Every rule must be on-chain. If the ledger has hidden liabilities, the monetary policy cannot be trusted. The satoshi count may be correct, but the complete set of conditions governing those satoshis is no longer visible.
SegWit broke the pricing. Taproot broke the audit trail. The monetary policy is sound. The settlement policy got debased. The ledger has a shadow.
The fix is simple. Not easy. Simple.
1 bit = 1 bit. Every byte in a block pays the same fee. No witness discount. No data-type distinctions. Every byte costs the same to store, validate, and serve. Every byte should cost the same to write.
This eliminates the asymmetry that makes gaming possible. You cannot exploit a pricing scheme with no differential pricing. A byte is a byte is a byte.
The thermodynamic argument: a byte on disk is a byte on disk. The network does not know the difference between a signature and an inscription. It stores both, validates both, serves both, forever. Charging different prices for the same thermodynamic cost is the externality. Removing the differential closes the gap.
The economic argument: if you want sound money, you need sound settlement. Sound settlement means the cost of writing to the chain reflects the cost of carrying the chain forward. The fee market must price persistence, not just inclusion. The writer must pay what the network actually costs.
The fix is not a new protocol. It is the removal of a distortion. Remove the discount. Let the fee market price all bytes equally. Let finality cost what finality costs.
The deeper point is not about SegWit or Taproot. It is about what happens when a system that guarantees permanence does not price it.
Bitcoin guarantees that every byte it stores will be stored forever. This is its most expensive property. It is the property that makes settlement final. It is the property that makes the system trustworthy. And it is the property that has not been priced.
The fee market prices the act of writing. It does not price the consequence of permanence. A writer pays a fee. The network carries the bytes forever. The gap between the fee and the cost of forever is the externality. It is paid by node runners. It is paid in perpetuity. It is the hidden subsidy at the heart of the settlement layer.
21 million is enforced by code. Block space was enforced by scarcity. Then the scarcity got inflated, discounted, and made opaque. The monetary policy is sound. The settlement policy got debased.
The question is not whether Bitcoin is sound money. The question is whether its settlement layer is.