crvUSD is a stablecoin with no “emissions story” to hide behind

crvUSD grows when borrowers take on debt and shrinks when they repay. That sounds obvious. The important part is what it implies for sustainability. For an emissions-driven contrast, see our Olympus tokenomics review.

There is no airdrop schedule. No liquidity mining for the stablecoin itself. No inflation knob to paper over weak product-market fit. Supply expansion has to be “earned” the hard way, by users posting collateral and choosing to hold a liability denominated in crvUSD because the pricing, liquidity, and liquidation behavior are good enough. Curve’s developer docs describe an overcollateralized stablecoin where users mint against a selection of crypto collaterals, and new collateral listings require DAO approval.

The mechanism-level differentiator is LLAMMA. Curve describes it as an AMM-based liquidation system that continuously rebalances collateral across a liquidation range rather than triggering a single liquidation point. In practice, this is the “real tokenomics” of crvUSD. It determines how much bad debt the system tends to create during stress, which determines whether the stablecoin can scale without subsidized liquidity forever.

Supply is elastic, but governed through debt ceilings and “pre-minted” inventory

crvUSD issuance is constrained by market-level debt ceilings. The ControllerFactory can set a market’s debt ceiling, and that action mints crvUSD to the market’s controller address when the ceiling increases, and burns from it when the ceiling decreases (subject to available balance). This is a very specific supply-control posture. It looks less like “mint per loan” and more like “authorize inventory, then let lending draw it down.”

At the token contract level, Curve documents that crvUSD can only be minted by the token’s minter, which is the Factory, and that minting is tied to debt-ceiling actions when markets are added or ceilings are raised.

This design creates a recurring analytics trap. ERC-20 totalSupply can be meaningfully larger than what most people intuit as “in circulation,” because the system intentionally holds idle crvUSD in multiple places.

Curve’s StablecoinLens docs are unusually explicit here: their circulating supply definition excludes idle crvUSD in PegKeepers, unborrowed crvUSD in Controllers, and crvUSD allocated to the FlashLender, and instead sums PegKeeper debt plus total debt across Controllers.

If you want a market-facing snapshot, CoinGecko currently reports circulating supply = 288,987,739, total supply = 288,987,739, and max supply = ∞. Treat that as a convenience layer, not a protocol truth layer, given Curve’s “inflated totalSupply” warning.

If you’re comparing how stablecoins present supply and sustainability, our USX tokenomics review is a useful reference point.

From an emissions-sustainability lens, this is healthy. A stablecoin with infinite theoretical max supply can still be sustainable if marginal issuance is continuously “paid for” by productive demand for leverage and settlement, and if contraction is credible under stress.

LLAMMA turns liquidation into a continuous market microstructure problem

LLAMMA is where Curve spends its complexity budget. It is also where crvUSD’s long-run scalability either becomes real or fails quietly.

Curve’s LLAMMA explainer frames liquidation as a mechanism with a range, not a single cliff: collateral is deposited into discrete bands, and users choose the number of bands when opening a loan, with a minimum of 4 and a maximum of 50.

Mechanically:

When collateral price falls into the liquidation range, the system begins selling collateral for crvUSD (soft-liquidation). When price recovers, the system can convert crvUSD back into collateral (de-liquidation).

Liquidation is not triggered “per account” in the typical liquidator-auction sense. Curve emphasizes it is executed on a band basis, which is a scalability choice. It concentrates execution into AMM state transitions that arbitrageurs are incentivized to keep aligned with an oracle price.

This is where tokenomics meets microstructure. The peg and solvency properties of crvUSD lean heavily on external actors:

In other words, crvUSD stability is not primarily “backing ratio theater.” It is throughput. Can the system continuously clear risk by trading, without creating bad debt, across real volatility regimes.

PegKeepers and monetary policy: automated balance-sheet ops plus a rate reaction function

crvUSD uses two orthogonal levers to push price back toward $1: PegKeepers (balance-sheet operations into specific pools) and a monetary policy contract (interest rate control).

PegKeepers are documented as specialized contracts that hold a pre-minted supply of crvUSD used only for peg stabilization. Curve notes in its PegKeeper overview that if the pre-minted crvUSD is not deposited anywhere, it can and should be counted as out-of-circulation.

When crvUSD trades above $1, PegKeepers deposit crvUSD into their linked pool and receive LP tokens, increasing the pool’s crvUSD balance and pushing price down. When crvUSD trades below $1, they burn LP tokens and withdraw crvUSD (only to the extent they previously deposited), reducing pool crvUSD balance and pushing price up.

Curve also documents that the update function is permissionless and callers are rewarded with a caller share. That is a keeper incentive layer embedded directly into the peg machinery.

PegKeepers feed into the interest rate. Curve states PegKeeper debt influences the interest rate via a “DebtFraction” term, and points to the MonetaryPolicy implementation for details.

On the monetary policy side, Curve’s interest-rate mechanics describe a rate function that depends on an aggregated oracle price of crvUSD sourced from multiple Curve pools and on PegKeeper debt relative to total system debt, alongside parameters like sigma, rate0, and target_debt_fraction.

The baseline parameter rate0 is constrained by a MAX_RATE equal to 400% APY, per Curve’s documentation of the contract constants and initialization checks.

Conceptually, Curve’s public-facing narrative for this system is consistent: the rate reaction function is designed to increase borrowing cost when the peg is weak and reduce it when the peg is strong, so debt supply responds endogenously.

As an emissions sustainability analyst, I read this as a hard commitment to productivity-linked expansion. The system can incentivize contraction via high rates, but sustainable growth requires a real reason to borrow and hold crvUSD at scale, net of that carry.

Fiscal flows: interest is the main “revenue,” and Curve routes it with on-chain plumbing

crvUSD does not tax holders via inflation. It taxes borrowers via interest and fees, then routes that value through Curve’s fee infrastructure.

At the market level, Curve documents a collect_fees() function on Controllers that collects borrowing-based fees and AMM-based fees, and routes them to a fee receiver specified in the Factory.

The Factory also defines fee_receiver as the receiver of interest and admin fees, and it can be updated by the Factory admin.

Curve’s fee plumbing then routes accumulated crvUSD fees onward to other contracts according to predetermined weights, with permissionless dispatch.

Downstream, the FeeDistributor is the contract that distributes the fee token to veCRV holders. Curve notes there are two FeeDistributors because fees were historically distributed in 3CRV until a DAO vote changed the reward token to crvUSD.

scrvUSD is the most direct “tokenomics productization” of these flows: Curve describes scrvUSD as a yield-bearing version of crvUSD built on a Yearn V3 vault, earning a portion of yield generated by borrowers paying interest to mint crvUSD.

The scrvUSD implementation makes the distribution rule explicit. Curve’s RewardsHandler docs state that weight() is the time-weighted average ratio of crvUSD deposited in the vault relative to circulating supply, expressed in basis points of total rewards requested, and that a minimum_weight lower bound can be used for bootstrapping and manipulation resistance.

For a lending-first comparison point on fee-driven token design, see our Maple Finance review.

Curve positioned scrvUSD as a scaling engine for crvUSD in its November 13, 2024 announcement, describing a loop where scrvUSD deposits stabilize the peg and lower borrowing rates, attracting more borrowers and increasing fee-derived yields. In Curve’s 2024 report, Curve states scrvUSD launched on October 31, 2024 and that a “Fee Switch” redirected a portion of fees from crvUSD markets to scrvUSD holders, with the DAO later increasing the share of revenues directed to scrvUSD holders.

This is a coherent fiscal design. It also creates a measurable sustainability constraint: scrvUSD yield is only “real” to the extent that crvUSD borrowing demand is real, because the yield is sourced from borrower-paid interest, not emissions.

Governance and parameter control: Curve can change the rules, and it matters for peg credibility

Curve’s documentation is clear that core expansion actions are governance-gated.

The Factory docs state that adding a new crvUSD market can be done by the CurveOwnershipAgent, and therefore requires a successfully passed DAO vote. The same Factory is also where debt ceilings and the fee receiver are controlled by the admin.

The MonetaryPolicy docs similarly frame admin control as CurveOwnershipAgent-controlled, including the ability to set rate0 (within the MAX_RATE constraint).

PegKeepers themselves evolved through governance pressure. Curve documents that PegKeeperV1 had issues including spam susceptibility and a depeg scenario where the paired stablecoin depegs and leaves unbacked crvUSD in circulation, and that PegKeeperV2 introduced a separate PegKeeperRegulator with pause controls and oracle-based guardrails like price_deviation and worst_price_threshold.

Concrete governance actions show how “monetary plumbing” changes over time. For example, Curve’s October 29, 2024 post states the Curve DAO approved and implemented a proposal executed on October 23, 2024 to integrate Mountain Protocol’s USDM into the PegKeeper system and set a 10,000,000 crvUSD debt ceiling for that PegKeeper.

The biggest governance-sensitive tokenomics shift is that Curve can authorize new issuance channels that sit outside the “borrow against collateral” loop.

On November 25, 2025, a governance forum proposal describes “new mint markets” as a parallel issuance channel for crvUSD that operates outside the PegKeeper mint-burn stabilisation loop, citing DAO-approved credit lines including 60,000,000 crvUSD for Yield Basis (effective September 24, 2025), expanded to 300,000,000 crvUSD on October 14, 2025, and a stated total of 315,000,000 crvUSD in credit lines outside the PegKeeper framework.

As a long-horizon observer, I treat this as the key governance risk surface. When a stablecoin can be expanded by policy decisions that are not mechanically forced by borrower demand, the system starts to resemble a discretionary central bank. That can work. It just changes what “credibility” means.

If you’re doing tokenomics consulting around crvUSD integrations, the practical work is stress-testing how parameter shifts propagate into peg, liquidity, and borrower behavior, then deciding which risks you can actually own operationally.

Risk: the peg is only as good as liquidation throughput (and discretionary issuance discipline)

crvUSD’s design is intellectually honest. It admits that stablecoin pegs are enforced through incentives, trading, and policy. Not vibes.

That also makes the failure modes legible. They cluster around (1) execution under stress and (2) governance decisions that change the monetary envelope.

Dominant risk: rapid, correlated stress that overwhelms LLAMMA + PegKeeper execution and forces “policy” to fight market microstructure.

LLAMMA’s promise is continuous liquidation. Continuous liquidation requires continuous counterparties. Curve explicitly relies on arbitrage incentives created by divergences between get_p and the oracle price, so arbitrageurs trade and rebalance collateral in bands. PegKeepers then perform pool-level balance-sheet ops based on whether crvUSD is above or below $1, and anyone can call update for a reward.

In normal conditions, that is elegant. Under abrupt regime shifts, it is fragile in a specific way: it depends on external trading appetite at the exact moment everyone’s risk budget is shrinking. If volatility spikes while liquidity thins, the cost of arbitrage rises, and “continuous liquidation” can become “continuous slippage.” Losses from soft- and de-liquidation and interest accrual are described by Curve as factors that reduce loan health, with losses depending on number of bands, liquidity, and arbitrage efficiency.

Here is the uncomfortable part. Peg stability then gets pushed onto the interest rate reaction function. Curve’s MonetaryPolicy formula responds to the aggregated crvUSD price and to PegKeeper debt relative to total debt, with parameters like rate0 and target_debt_fraction. That can raise rates aggressively, up to the documented maximum, which is itself extremely high.

High rates can force contraction by making debt expensive to carry. They can also break the growth engine by scaring off marginal borrowers for long periods, which reduces “productive” demand and undermines the very fee flows that scrvUSD and veCRV depend on. This is the macro trade-off in crvUSD: short-term peg defense via punitive carry versus long-term equilibrium where borrowing is competitive and stable demand exists.

The governance forum proposal from November 25, 2025 effectively documents this tension in the wild, describing rate volatility and peg volatility when debt composition shifts and when credit lines outside the PegKeeper framework expand.

If you want one metric to watch that captures this dominant risk, it is the joint behavior of (a) crvUSD price deviation, (b) PegKeeper debt share, and (c) the policy rate trajectory. Curve’s own docs tie all three together in the rate equation, and you can track similar indicators over time in our research reports.

It is also worth noticing that Curve itself documents prior PegKeeper design issues around depegging of the paired stablecoin leaving unbacked crvUSD in circulation, which is exactly the kind of tail event that shows up during systemic stress. V2 adds mitigations, but mitigations do not remove tail risk.

From a sustainability standpoint, the lesson is simple. crvUSD can scale without emissions. It cannot scale without resilient liquidation throughput and disciplined discretionary issuance. Those are harder.

Top 3 risks

  1. Liquidation throughput failure under fast drawdowns. Trigger: sudden collateral price gaps that traverse multiple LLAMMA bands before arbitrage rebalances. Mechanism: soft-liquidation and de-liquidation depend on arbitrage between get_p and the oracle price, and Curve notes losses depend on liquidity and arbitrage efficiency, which can degrade under stress. Who bears it: borrowers (collateral converted at adverse execution), and second-order crvUSD holders if undercollateralization pressure leaks into peg weakness. Measurable indicators: worsening crvUSD price deviation, rising hard-liquidations, and sustained elevation in the MonetaryPolicy rate.

  2. Discretionary supply expansion via credit lines (“mint markets”) destabilizes the peg regime. Trigger: large DAO-authorized credit lines outside the PegKeeper framework expand faster than organic borrower demand. Mechanism: supply can grow in a channel not mechanically tied to collateralized borrowing and the PegKeeper loop, changing debt composition and stressing the monetary policy feedback, as described in Curve governance discussion. Who bears it: crvUSD holders (peg volatility), scrvUSD depositors (yield variability driven by unstable borrow demand), and the DAO (credibility cost). Measurable indicators: rising authorized credit capacity, higher peg volatility, and “erratic” PegKeeper activity described by proposers.

  3. PegKeeper paired-stablecoin risk and oracle-guardrail edge cases. Trigger: significant depeg or credibility shock in a stablecoin paired with crvUSD in PegKeeper pools. Mechanism: Curve documents a PegKeeperV1 scenario where a paired coin depegs and the PegKeeper can be left with debt and unbacked crvUSD in circulation; while V2 introduces oracle verification and worst-price thresholds, the system remains exposed to multi-asset correlated failures. Who bears it: crvUSD holders (peg risk), and LPs in affected pools (inventory risk). Measurable indicators: PegKeeper deposit halts due to guardrails, widening price divergence between pool price and oracle EMA, and reduced PegKeeper debt capacity utilization.

crvUSD is one of the cleaner attempts at a non-emissions-backed stablecoin scale strategy. The design pays for itself only when the system’s leverage demand is real and when governance treats discretionary issuance as a last resort rather than a growth hack.



This article is part of our Tokenomics Deep Dive series.