Stablecoins and Settlement Risk

The short answer

A stablecoin is a token engineered to trade at or near a reference price, usually a national currency. It is not stable by nature; it is stable while a mechanism keeps it there.

That mechanism is the entire subject. Behind every stablecoin sits a claim on something — reserves, collateral, or nothing but an incentive scheme — and the risk you are taking is the risk that the claim does not hold when it matters.

What a stablecoin structurally is

Strip away the naming and a stablecoin has three parts.

1. A token. A transferable unit on a network, which moves like any other token and settles like any other token.
2. A reference price. The number it is supposed to track. Usually one unit of a national currency.
3. A stabilisation mechanism. Whatever is supposed to close the gap when the market price and the reference price separate.

The first two are visible to anyone. The third is where all the variation lives, and it is the part most holders never inspect.

Note what is absent from that list: any legal rule that the price must equal the reference. Nothing enforces the peg. The peg exists because someone can profit from closing the gap, or because someone is obliged to redeem at par, or both. Remove the profit or the obligation and the price does what any other traded price does.

The backing models

The categories below describe designs, not products. Any real token should be read against them, not assumed into them.

Two observations that apply across all four.

The backing and the claim are different things. Assets can exist and still be unavailable to you. What matters is whether you, or somebody whose arbitrage benefits you, can convert tokens into the reference asset at par, promptly, in size.

An attestation is a snapshot. It reports something about a moment in time, usually about assets and often less about liabilities and encumbrances. The same limitation covered in custody and exchange risk applies here: point-in-time evidence of assets is not evidence of solvency.

Why "stable" is a claim, not a property

Volatility is a measurement of the past. A quiet history is compatible with a mechanism that has simply not been tested yet — and stabilisation mechanisms are specifically the kind of thing that works until the day a large enough number of holders want out simultaneously.

The reflexive part is worth sitting with. Most of these designs rely on arbitrage: if the token trades below the reference, someone buys it cheaply and redeems it at par, and that buying closes the gap. This depends on the arbitrageur believing redemption will be honoured. Doubt about redemption removes the buyer, and removing the buyer is exactly what widens the gap that caused the doubt.

That is why a peg tends not to erode smoothly. It holds, and then it does not.

What a depeg means for a position

The subtle point: a stablecoin is usually your denominator, not your holding. It is the unit you quote in, the unit you post as collateral, and the unit you settle in. A move in the denominator therefore hits you in several places at once.

The numbers below are invented for teaching. They describe no real token and are not a trade idea.

Suppose a stablecoin referencing one dollar trades at $0.96, and you hold a spot position quoted in that stablecoin.

• Quoted price. The pair may rise while the value of your holding in dollars is unchanged, or even falls. The chart tells you less than it appears to.
• Collateral. If the same token margins a leveraged position, the venue may revalue it, tighten requirements, or leave the valuation stale — each of which produces a different, unpleasant surprise. Spot vs derivatives in crypto covers what happens when margin moves against you.
• The exit. Converting out costs the discount plus a spread that has widened precisely because everyone else is trying the same thing. Withdrawal queues and paused conversions have appeared in past stress episodes across the sector.
• The hedge that was not one. Traders park in stablecoins to be flat. In a depeg you are not flat. You hold a position in an instrument with its own risk.

None of this is directional. You can be entirely correct about the asset you traded and still be worse off.

The questions worth asking

Before you hold any stablecoin in size, or margin anything with one:

• What is the claim, and against whom? Is there an identified issuer, an on-chain contract, or neither?
• Can it be redeemed at par? By whom — anyone, or only approved counterparties with a minimum size?
• What sits in reserve or collateral, and how liquid is it on a bad day rather than an average one?
• Who verifies that, how often, and what exactly do they verify?
• What happens mechanically if the price separates from the reference? Who is supposed to close the gap, and what do they need to believe in order to do it?
• Where is it held? On a venue, it is a claim on a claim — an issuer claim, held for you by an exchange.
• What is your jurisdiction's treatment of it, and does that change whether anyone stands behind it?

If you cannot answer the first two, you do not know what you are holding.

What to do with this

Treat stablecoin balances as a position with a counterparty, not as cash. Size them like a position: ask what you would lose if that particular claim were worth materially less than par, and decide whether that number is acceptable.

Keep a second denominator. Track account equity in the currency you actually spend as well as in whatever the venue quotes, so a move in the unit itself cannot hide inside your P&L.

And keep the settlement question separate from the trading question. Whether an asset goes up is one problem. Whether the thing you are paid in holds its value, and whether you can convert it when you want to, is a different problem — and it does not go away by being right about the first one.