EVMORE vs PAXG: bearer asset versus custodial claim

PAX Gold (PAXG) and EVMORE end up next to each other in a lot of searches. The phrase “digital gold on Ethereum” returns both. Both are ERC-20s. Both pitch themselves around scarcity. A buyer skimming descriptions could be forgiven for treating them as substitutes.

They are not substitutes. They are different categories of instrument, and the difference — bearer asset versus custodial claim — matters for almost every reason you might buy either.

What PAXG is

PAX Gold is an ERC-20 issued by Paxos Trust Company. Each token represents one fine troy ounce of London Good Delivery gold held in Brink’s vaults in London. The token is fully backed; Paxos publishes attestations from a third-party accounting firm; Paxos is regulated by the New York Department of Financial Services.

The instrument’s mechanics:

• Mint. A KYC’d Paxos customer wires fiat, Paxos buys gold, custodian allocates the gold to the vault, Paxos mints PAXG.
• Burn. A KYC’d Paxos customer burns PAXG, Paxos un-allocates the gold, customer receives gold (above thresholds) or fiat equivalent.
• Transfer. Standard ERC-20 transfer, with a small onchain fee schedule.

PAXG is an excellent vehicle for what it does: tokenised exposure to physical gold inside Ethereum. The price tracks spot gold closely. The custody is regulated. The accounting is audited.

It is also, structurally, a claim. The token is the receipt; the gold is the underlying. Bearer of the token has a claim on the underlying as long as Paxos is operating, solvent, regulated, and willing to redeem.

What EVMORE is

EVMORE is an ERC-20 whose contract enforces a 21-million supply cap and a halving issuance schedule. Tokens enter circulation only through verified KeccakCollision mining proofs submitted on-chain. There is no custodian. There is no underlying asset to redeem against. There is no Paxos.

The instrument’s mechanics:

• Mint. Anyone submits a valid 128-byte mining proof to EvmoreToken.vy. The 62-line verifier confirms the proof. The contract mints the era’s block reward to the submitter.
• Burn. Optional burn() function for accounting. Tokens removed from supply forever.
• Transfer. Standard ERC-20 transfer, no admin pause, no blocklist.

EVMORE is a bearer asset. The token is the asset. There is nothing the token is a claim on.

The bearer-vs-claim distinction

In traditional finance, “bearer” and “claim” describe two very different things:

• A bearer asset is one whose ownership is settled by holding it. Physical cash, gold bars, physical bond certificates in the old days. The bearer is the owner.
• A claim is a receipt for something else. A bank deposit, a brokerage position, a vault receipt, a tokenised commodity. The bearer holds a right against an issuer.

Bearer assets fail when you lose the asset. Claims fail when the issuer fails.

The history of financial markets is, to a first approximation, the story of how claims have become more reliable than bearer assets through institutional infrastructure (banking systems, custodians, clearing houses, regulators). For most users in most jurisdictions, a claim against a regulated custodian is now safer than a bearer asset stored at home.

Crypto inverts this. The bearer asset becomes safer than the claim again, because the technology lets you self-custody at scale. The friction of holding cash safely was high (vaults, security guards, insurance); the friction of holding a private key is, modulo seed-phrase management, low.

PAXG operates on the traditional logic: it is a claim, backed by a regulated custodian, with the safety profile of the custodian. EVMORE operates on the crypto-native logic: it is a bearer asset whose security profile is the contract code.

Where the difference manifests

The bearer-vs-claim distinction manifests in five places that matter:

1. Self-custody. PAXG can be self-custodied in the sense that you can hold the ERC-20 in your own wallet. But the underlying gold cannot be self-custodied; it is in a vault in London. Redemption requires you to become a Paxos customer, satisfy KYC, and meet minimum thresholds. EVMORE is self-custodied end-to-end: the token in your wallet is the asset; there is no “underlying” elsewhere.

2. Censorship resistance. PAXG has a pause() capability. The token can be frozen at the contract level. Specific addresses can be blocklisted (this is a Paxos compliance function, used at the direction of regulators or law enforcement). EVMORE has neither pause nor blocklist; transfers cannot be administratively blocked.

3. Counterparty risk. PAXG inherits Paxos as a counterparty. If Paxos becomes insolvent, the bankruptcy process determines what happens to PAXG holders — a process that, even in well-regulated jurisdictions, is messy and slow. EVMORE has no counterparty; the contract code is the entire dependency surface.

4. Operational risk. PAXG depends on Brink’s continuing to operate, on the periodic attestation pipeline functioning, on Paxos’s banking relationships being intact, on the vault’s physical security holding. EVMORE depends on Ethereum continuing to operate and the Vyper contract continuing to behave as written. The dependency lists are different.

5. Regulatory exposure. PAXG is unambiguously a regulated instrument under multiple regimes. Its holders inherit the regulatory environment of the issuer. EVMORE is software; it has no issuer to regulate. Holders are subject to whatever regulations apply to their general crypto holdings in their jurisdiction, but there is no issuer-level exposure they are inheriting.

When PAXG is the better instrument

I want to be careful here. PAXG is not “wrong.” It is a different tool, and there are use cases where it is the right tool:

• You want exposure to spot gold price. PAXG tracks spot gold within typical premium. EVMORE has its own price discovery; it does not track gold.
• You require regulatory cover. Holding PAXG in a regulated entity (a fund, a corporate treasury under audit) is straightforward because PAXG is a regulated instrument. EVMORE may not satisfy compliance for some institutional buyers.
• You want institutional custody. Major institutional custodians (Anchorage, BitGo Trust, Fidelity Digital Assets) have established custody for PAXG. EVMORE is new; institutional custody offerings will take time.
• You want auditable physical backing. “There are X tonnes of gold in a London vault, attested quarterly” is meaningful to some buyers. EVMORE has no analogous claim; there is no physical backing because there is no claim.

For these buyers, PAXG is correct.

When EVMORE is the better instrument

For other buyers, EVMORE is correct:

• You want a bearer asset with no issuer to regulate. EVMORE has no issuer.
• You want native DeFi composability without custody risk. EVMORE is born on the EVM; PAXG inherits Paxos’s custody risk.
• You want self-custody that does not bottom out at a vault. EVMORE’s “underlying” is the contract; PAXG’s underlying is in Brink’s.
• You want monetary policy enforced by code, not by attestation. EVMORE’s supply cap is in Vyper; PAXG’s supply is whatever Paxos has minted.
• You want exposure to digital scarcity, not gold spot. EVMORE’s price discovery is its own; PAXG’s is gold’s.

These buyer profiles are different. The instruments are different. The category overlap is largely an artefact of the “digital gold” search query.

The composition question

The composability story is starker than the spot-buyer story.

A DeFi protocol that accepts PAXG as collateral is implicitly accepting Paxos as a counterparty. Liquidation logic must price the custodian risk. Lending markets typically apply a “custodian-risk haircut” to wrapped or tokenised commodities, which is a polite way of saying users pay for the custodian.

A DeFi protocol that accepts EVMORE as collateral is accepting EVMORE — a smart contract on Ethereum — as collateral. The risk to model is the contract risk and the asset’s price volatility. There is no off-chain counterparty to haircut for.

This is part of why EVMORE-style native ERC-20 scarcity assets are structurally interesting for DeFi protocol designers. The risk model is contained inside the chain. Nothing is being imported from off-chain reality.

The verdict

PAXG and EVMORE share a shelf in the metaphorical store. They do not share a customer.

If your question is “how do I get tokenised gold exposure with regulated custody?” the answer is PAXG. If your question is “how do I get bearer digital scarcity with no issuer?” the answer is EVMORE. They are not substitutes.

Treating them as substitutes is how people end up disappointed in both.