One month after the Usual.Money depeg: The need for a new DeFi primitive in long-dated liquidity
Source: https://leviathannews.substack.com/p/collateral-damage-usd0-depeg-leaves?r=1yj9r&utm_campaign=post&utm_medium=email&triedRedirect=true
On January 9, 2025, DeFi ran into yet another liquidity crisis. This time, it was USD0++, a staked version of USD0, that depegged—slipping from $1 to as low as $0.87 in a matter of hours. The problem wasn’t new, but the way it happened made it clear that DeFi still hasn’t figured out how to handle liquidity for long-term assets.
At first glance, USD0++ looked simple: stake your USD0 for four years for a single maturity, earn extra yield + points + governance. Debt-like risk.. equity-like upside → payout is similar to a convertible… right?
But when investors needed to exit early, they realized there was no real way to do it— no natural bid-ask buyers, just a steep discount on secondary markets on AMMs. It’s almost as if everyone assumed it is always at par because of the AMM nature of most DeFi designs. The problem was, in this design, that liquidity itself had no price.
The real problem wasn’t “net asset value,” it was the price of “net liquidity value”
USD0++ is being treated here as a net asset with a slapped on “discount rate.” Yes, this is very finance 101, “time value of money” and technically the correct synchronous “price.”
$$
\text{Price} = \text{NAV} \times (1 - \text{discount}) $$
But let’s rethink how we define “discounted value,” but instead of the classic discount rate why not treat the discount rate as a market-driven input, representing the true cost of liquidity?
Right now, USD0++ are often assumed to be worth 87 cents on the dollar just because that’s the “net asset value” (NAV). IMO that logic is skeuomorphic—a carryover from TradFi, where NAV assumes orderly markets and predictable liquidity. DeFi doesn’t work that way.
$$ U = \frac{\text{Total Borrowed}}{\text{Total Liquidity}} $$
Aave/Compound/other floating rate money markets already use a different approach. Instead of relying on static inputs, they treat utilization as the core control variable for the kinked curve—a feedback loop where interest rates adjust dynamically based on available liquidity. The system self-regulates with a “P-controller,” auto-calibrating loan-to-value (LTV) ratios as utilization approaches a predefined threshold.
For long-duration assets, however, utilization isn’t the right input— “fixed-income” instruments don’t function like revolving credit (it’s literally called “fixed” income, you get a defined yield). Instead, the key variable must be the liquidity itself. Rather than assuming a fixed discount rate for early exits, the market should determine the cost of liquidity dynamically & diachronically, just as Aave allows interest rates to float in response to borrowing demand. Relying on secondary market liquidity for price discovery is too fragile; the redemption queue mechanism must be adjusted at the protocol level to prevent reflexive depegs.
This shift transforms the discount rate from a rigid assumption into an actively priced variable, making long-duration DeFi assets more flexible and capital-efficient. Historically, liquidity pricing in DeFi has been structured around utilization-based models, pioneered by Aave and similar protocols. These approaches have worked well for short-term lending but fail to account for liquidity mismatches in fixed-maturity assets.
<aside>
Fixed-income DeFi risks repeating the mistakes of the regional banking crisis in 2023, where banks held long-term assets under the assumption that they could always wait for maturity. When depositors began withdrawing funds en masse, banks had no choice but to sell at steep discounts, not because the assets were bad, but because liquidity wasn’t priced correctly. It wasn’t a credit crisis—it was a liquidity mismatch.
</aside>
$$ P_{\text{exit}} = P_{\text{NAV}} \times (1 - D(L)) $$