For most tokenized securities, issue on an Ethereum Layer 2: you inherit Ethereum’s settlement assurance at lower cost, keep EVM tooling and compliance-at-transfer support, and reach where investors and stablecoins already are. Permissioned chains are for mandates that require them.
The chain is where your ownership record lives for the life of the asset. Pick it for settlement assurance and longevity, not for a launch-week gas quote.
Why
A tokenized security has one job the chain must protect: keep a correct, enforceable record of who owns what, for years, through counterparties you don’t control. Five things decide whether a chain does that well.
Settlement assurance. Ethereum mainnet has the strongest guarantee that a settled transfer stays settled — a large, decentralized validator set and economic finality after two epochs, roughly 13 minutes. (Ethereum.org — finality) That assurance is the product. Everything below is a way to get most of it for less money.
What a rollup actually inherits. Arbitrum and Base are optimistic rollups: they execute transactions off mainnet and post the data and proofs back to Ethereum, so data availability and dispute settlement anchor to L1. (Arbitrum docs, Base docs, Ethereum.org — L2) You get Ethereum-anchored settlement at cents per transfer. The honest caveat: today most rollups still run a single sequencer and enforce a challenge window on withdrawals to L1 — track the specific chain’s maturity rather than assuming “L2” means one thing. (L2BEAT)
Where Polygon is different. Polygon PoS is a commit-chain with its own validator set; it periodically checkpoints to Ethereum but does not derive its security from L1 the way a rollup does. (Polygon docs) It has deep liquidity, a large user base, and very low fees, and for some issuers that reach matters more than settlement purity. Say that plainly — but know you are choosing a different security model, not a cheaper Ethereum.
Where your investors and money already are. EVM chains share tooling: the compliance token standards (ERC-3643, ERC-7943), qualified-custodian support, wallets, and — decisively — regulated stablecoins for subscription and distribution. USDC and USDT are native on Ethereum and the major L2s. A chain with no stablecoin and no custodian is a worse record even if it is technically elegant.
Cost and longevity. Mainnet is expensive per transfer; L2s are cheap. But cost is the easy variable. The hard one is whether the chain is credibly neutral and will still be maintained a decade from now, when the asset is still outstanding and someone needs to enforce a transfer.
The edge cases
When mainnet L1 is right. High-value, low-frequency instruments — a single large bond, a fund whose transfers are rare — where per-transfer cost is rounding error and you want the maximum settlement guarantee with no L2 caveats.
When a permissioned chain is right. A mandate that requires it: a central-bank or wholesale-settlement project on a permissioned ledger, or a deal that must settle atomically against tokenized deposits or a wholesale CBDC on specific infrastructure. Here the infrastructure dictates the chain, and a public L2 is simply not eligible. The trade you accept: no public liquidity, no shared stablecoins, no neutral composability, and responsibility for the chain’s own liveness.
When Polygon PoS is right. Consumer-scale, high-frequency, lower-ticket distribution where fee and existing user base dominate and the different security model is acceptable to your counsel and your investors.
The trap. Choosing a chain because a vendor only supports that chain. The chain should follow the asset and the investors; if the platform forces the chain, that is a platform limitation wearing an architecture argument.
What this means for your structure
Default to an Ethereum Layer 2 unless a specific reason moves you off it. It gives you Ethereum-anchored settlement, EVM compliance tooling, native regulated stablecoins, and a cost structure that does not punish ordinary transfers.
Stobox Compass issues security tokens primarily on Base, an Ethereum Layer 2, for exactly these reasons: Ethereum settlement, standard EVM tooling for compliance-at-transfer, and USDC-native rails. It also issues on Arbitrum, on Canton for institutional mandates that need a permissioned ledger, and on other EVM chains. The chain follows the asset and the mandate — decide it with counsel before you pick a platform, not after.
Whatever you choose, the test is the same one that decides everything else in tokenization: if the platform disappears, does the record survive? A chain with strong settlement assurance and open tooling passes it. A chain chosen for launch-week convenience often does not.
Gene Deyev’s take
Founders obsess over launch-week gas fees, which is the one variable that stops mattering the day after you issue. The question that actually matters is whether you can leave: can you move the ownership record off this chain, in a decade, without asking a vendor’s permission? That is why I default to an Ethereum L2 with open compliance standards over anything a single platform locks you into. A cheap chain you can’t exit is more expensive than a costly one you can.
— Gene Deyev, CEO & Co-Founder, Stobox. Co-author of the Stobox Tokenization Framework and the STV3 protocol; ERC-7943 backer; SEC Crypto Task Force roundtable participant (2025).
Related questions
- Tokenizing through an SPV — the structure the chain sits under.
- What if the platform disappears? — why settlement assurance and open standards are the real selection criteria.
- ERC-3643 vs ERC-7943 — the compliance-at-transfer standards that ride on the chain you pick.
Last updated: 2026-07-12.