Half a percent is a strange discount for something that claims to be a dollar. That was the gap a friend showed me between his stablecoin balance on Avalanche and the same stablecoin on Ethereum, same aggregator, same moment. The explanation took longer than either of us expected, because the two balances were different assets with different issuers, different backing, and different failure modes. One was USDC, issued by Circle, a claim on Circle's reserves. The other was USDC.e, a receipt issued by a bridge, a claim on whatever that bridge still holds. His wallet displayed both as more or less the same thing.
This trips up people who are otherwise careful. The entire pitch of a stablecoin is fungibility, a dollar is a dollar wherever it sits, so the idea that two tokens with nearly identical names on adjacent chains could carry meaningfully different risk feels like a technicality. It stops feeling like one the first time a bridge halts withdrawals.
What native actually means
A native deployment is one where the issuer itself put the token contract on that chain. Circle deploys USDC natively on Ethereum, Solana, Base, Arbitrum, Avalanche and a growing list of others. Tether does the equivalent for USDT on the chains it supports. On a native deployment the issuer controls minting, the tokens are a direct claim on the issuer's reserves, and an institution with an account at the issuer can redeem them for actual dollars. The chain is just a ledger choice. The liability sits with the issuer either way.
Bridged versions come from a completely different process. Early in a chain's life, before the issuer has shown up, people still want dollars on it, so a bridge steps in. You send native USDC to the bridge's contract on Ethereum, the contract locks it, and the bridge mints a token on the destination chain. That minted token is an IOU against the locked pile. Circle never touched the destination chain. Circle's obligation ends at the tokens sitting in the bridge contract, and if you show up at Circle's redemption desk holding the bridged version, they owe you nothing, because they did not issue it.
The naming makes this easy to miss. The .e suffix on Avalanche originally meant bridged from Ethereum. Arbitrum spent its early period with a bridged token that was simply called USDC, and when Circle later launched a native deployment there, the old bridged version got relabeled USDC.e after the fact. Polygon went through the same shuffle. So the same suffix carries slightly different history on different chains, plenty of frontends drop the suffix entirely, and nothing stops anyone from deploying an unrelated token with the same name. The symbol is a courtesy with no enforcement behind it.
What happens when a bridge fails
A lock-and-mint bridge is a custodian with smart contract keys, and custodians fail in a few well-documented ways. The contract gets exploited and the locked collateral is drained. The operators lose the keys, or turn out to have had unilateral control of them all along. Or the whole operation simply stops, which is what happened with Multichain, historically one of the larger bridges, when it abruptly halted and the bridged stablecoins on several chains it served slid to a deep discount. Holders still had a token that said USDC or USDT in the wallet. What they actually held was an unsecured claim on a bridge that had stopped answering.
The detail worth sitting with is that the underlying native tokens were fine the whole time. USDC on Ethereum did not wobble. The depeg lived entirely in the bridged layer, which makes sense once you see the bridged token for what it is, a credit instrument whose counterparty is the bridge. When Wormhole was exploited years back, the wrapped assets it had issued were only made whole because a deep-pocketed backer chose to refill the hole. Nothing in the design guaranteed that outcome, and there is no rule that says someone bails out your bridge.
There is a quieter failure mode too, and it is the one you are far more likely to hit. Liquidity walks away. Once an issuer launches a native deployment on a chain, market makers, DEX pools and lending markets migrate to it, because native is strictly better collateral. The bridged version keeps trading, but it thins out. Spreads widen, the swap route into native starts costing real basis points, lending markets deprecate it as collateral. If you are sitting on a large bridged balance through that transition, your exit gets more expensive every month, and nothing in the wallet UI tells you it is happening.
Burn-and-mint fixes most of this
Issuers know all of the above, which is why the newer transfer standards skip the collateral pool entirely. Circle's CCTP moves USDC between chains by burning native tokens on the source chain and minting native tokens on the destination chain, with Circle's attestation service confirming the burn before the mint goes through. There is no locked pile, no wrapped receipt, and no bridge counterparty holding your backing. What arrives on the other side is the issuer's own token, the same claim on the same reserves you started with.
Two caveats. First, burn-and-mint concentrates trust in the issuer's attestation infrastructure, so you have swapped bridge risk for a dependency you arguably already had, since the issuer could always freeze funds or mismanage reserves regardless of how the tokens moved. Second, routing frontends will sometimes quote you a route labeled with the official standard that actually hops through a pool of bridged tokens because it is a few basis points cheaper at that moment. The label on the button and the asset that lands in your wallet are separate questions, and only the second one matters.
How to check what you are actually holding
The check takes about two minutes per position and I would run it on any stablecoin balance large enough that losing a few percent of it would annoy you.
- Start from the contract address rather than the symbol. Click through to the token in a block explorer and copy the address your balance actually lives at.
- Compare it against the issuer's published list. Circle and Tether both maintain official pages listing the exact contract address for every chain they natively support. If your address matches, you hold the native asset. If it does not, you hold someone's wrapper, whatever the symbol says.
- If it is a wrapper, work out whose. The explorer will usually show the deployer or the address holding the minter role. A bridge contract in that role tells you exactly which organization's solvency you are exposed to.
- Check the exit. Find the deepest pool pairing the bridged version against the native one and estimate what converting your full balance would cost. That spread is the market's live estimate of the wrapper risk, and it is also roughly your exit price in a hurry.
- Recheck after any bridge incident and after any native launch on that chain, because both events move liquidity quickly.
My rough operating rule is that a native stablecoin is issuer risk, and a bridged stablecoin is issuer risk stacked on bridge risk, so it gets sized the way you would size a lower-grade credit. Holding a bridged token to LP a pool or farm an incentive is a reasonable, paid position with a known exposure. Parking dry powder in one because the wallet said USDC and you never looked closer is how people end up learning bridge mechanics on a bad day. The two minutes with a block explorer are a cheap way to avoid being one of them.