Bible Network Crypto DeFi Onchain RWA AI Agent Stablecoin Chain SAFU CryptoTax DeFAI AGI Claude Me Claude Skill Claude Design Claude Cowork
Independent Media
Not affiliated with any project
The Deepest Stablecoin Knowledge Base
stablecoin-bible.com
LATEST
GENIUS Act Is Now Law: What America's First Stablecoin Legislation Actually Changes — And What To Do With Your USDT and USDC  ·  Ethena USDe Complete Guide: How a Synthetic Dollar Is Born, How Delta-Neutral Generates Yield, sUSDe Mechanics and Insurance Fund Design  ·  Why Can Stablecoins Always Be Worth $1? Market Arbitrage, Direct Redemption, and PSM — Three Peg Defense Layers Explained  ·  How to Earn Yield on Stablecoins: sUSDS, sUSDe, Aave Lending, and CEX Earn Fully Compared  ·  MiCA Stablecoin Complete Guide: EMT Classification, How USDC Became Compliant, Why USDT Was Delisted, and USDS's Gray Zone  ·  Sky (ex-MakerDAO) and USDS Deep Dive: CDP Mechanism, PSM Peg Stability, RWA Reserves, and How SSR Makes USDS DeFi's Most Straightforward Yield Dollar
Glossary · stablecoin-types

Stablecoin Collateral Types

stablecoin-types 初級

30-Second Version · For the impatient
Why a stablecoin is worth $1 depends entirely on what backs it. The three main types today are: fiat cash (USDC, USDT), crypto over-collateralization (DAI, USDS), and synthetic derivatives strategies (USDe). Each collateral type determines the stablecoin's risk layers, yield potential, and resilience in extreme markets. Choosing a stablecoin is fundamentally choosing which type of 'guarantee mechanism' you're willing to trust.
Full Explanation +
01 · What is this?

Why do crypto over-collateralized stablecoins need 'over' collateralization — wouldn't 1:1 work?

Because crypto asset prices fluctuate — and fluctuate significantly.

Imagine: you lock $1,000 of ETH in and mint $1,000 of USDS. If ETH drops 30% tomorrow, your collateral is only worth $700, but the USDS you issued is still a $1,000 promise — the protocol has a $300 shortfall, and the stablecoin has lost its backing.

'Over-collateralization' is mandatory buffer: you lock in $1,500 of ETH but can only mint $1,000 of USDS (150% collateral ratio). ETH would need to drop over 33% before your collateral ratio falls below 100% — and before reaching 100%, the system triggers liquidation (typically at 130-150%), force-selling your ETH to repay, ensuring USDS always has adequate backing.

The required over-collateralization degree depends on collateral risk: ETH being relatively stable typically requires 150-200%; more volatile small-cap tokens may require 300%+; and USDC as collateral can be accepted at 110-120% (since it doesn't fluctuate itself).

02 · Why does it exist?

In a major market crash, which collateral type performs most stably?

There's no simple answer since each type has different vulnerability points under different stress scenarios.

Fiat-reserve type (USDC) is generally most stable during market crashes, because its collateral (dollars and short-term Treasuries) doesn't follow the crypto market. During March 2020 and the 2022 Terra collapse, USDC maintained prices very close to $1. The only exception was the 2023 SVB/Signature banking crisis — bank risk transmitted to USDC.

Crypto over-collateralized type (USDS) faces the most pressure during market crashes, because collateral (ETH, BTC) falls with the market, triggering mass liquidations simultaneously, and liquidation pressure may exceed system capacity, or network congestion may prevent timely liquidation (as in 2020's Black Thursday).

Synthetic type (USDe) faces unique stress during market crashes: during major crypto crashes, funding rates may go negative (because too many are short), making USDe yield disappear or even slightly negative. But since it holds a 'long-short hedged portfolio' rather than direct crypto assets, the principal itself doesn't shrink from market drops — as long as Ethena's short positions aren't force-closed by exchanges.

Conclusion: If you need the safest stablecoin during crypto market crashes, USDC is the top choice — but watch for bank partner risks.

03 · How does it affect your decisions?

What's the difference between RWA (Real World Assets) as stablecoin collateral and traditional fiat reserves?

These two are easily confused but have important differences.

Traditional fiat reserves: The issuer directly holds US dollar cash or bank deposits. Users can redeem directly from the issuer; the issuer uses cash in bank accounts to pay. The entire process occurs within the traditional financial system.

RWA (Real World Assets): In this context, RWA usually refers to US Treasury bills (T-bills) and similar real-world financial assets tokenized on blockchain. Sky Protocol purchases actual US Treasuries, 'tokenizes' ownership of those Treasuries on-chain as part of USDS's reserve, and distributes Treasury interest to USDS holders (via SSR).

Key differences:

  • Fiat reserves: cash → held in bank → high risk if bank fails
  • RWA: T-bills → managed on-chain → adds tokenization contract technical risk, but T-bills' credit quality is higher than bank deposits (US government credit vs commercial bank credit)

Practical meaning for holders: The RWA portion of USDS means you bear the risk of 'Sky Protocol correctly managing Treasury assets,' not the risk of any specific bank. In the 2023 Signature Bank failure, if USDS reserves were entirely RWA (Treasuries), the impact would be smaller than USDC's bank deposit model.

04 · What should you do?

What direction will stablecoin collateral evolve in the future?

Several trends worth watching:

1. Rising RWA proportion: As Fed rate cut expectations weaken and Treasury yields remain in the 4-5% range, incorporating RWA (T-bills) into stablecoin reserves both improves safety (US government credit) and creates real yield for holders. Sky's USDS and Ondo's OUSG represent this direction — expect RWA's share in DeFi stablecoin reserves to increase substantially over the next 3-5 years.

2. AI Agent payments driving new demands: As AI agents begin to autonomously complete tasks and payments, demand rises for 'no-human-approval required automated micropayment' stablecoins. These scenarios require minimal transaction costs and maximum interoperability, potentially catalyzing new lightweight stablecoin designs with more automated and intelligent collateral mechanisms.

3. Regulation pushing fiat-reserve type toward further standardization: Both GENIUS Act (US) and MiCA (EU) are pushing fiat-reserve stablecoin standards higher (HQLA, monthly audits). This will narrow quality differences among fiat-reserve stablecoins, but also make them more like 'digitized money market funds' — safe but yielding nothing.

4. Hybrid types continuing to evolve: The most competitive stablecoins in the future may be 'hybrid' — fiat base position + RWA yield + decentralized governance. This design tries to balance regulatory compliance, decentralized spirit, and yield capability, with USDS being the closest current example.

Real-World Example +

Same 'stablecoin' label, three very different collateral mechanisms during the May 2022 Terra collapse

In May 2022, Terra's algorithmic stablecoin UST collapsed from $1 to $0.02 in 72 hours, wiping out $40 billion. During the same period, let's see how stablecoins with different collateral mechanisms performed:

USDC (fiat-reserve type): Throughout the Terra collapse, USDC maintained an extremely narrow range of $0.998-$1.002. Reason: USDC's collateral (dollar cash and Treasuries) was completely unrelated to Terra's collapse — Luna crashing didn't affect Circle's dollars sitting in bank accounts. USDC holders felt almost no volatility during this crisis.

DAI (crypto over-collateralized type): DAI briefly showed a slight premium during the Terra collapse (up to $1.02) — because market panic drove large numbers of users into DAI as a safe haven, demand exceeding supply. DAI's mechanism had no problems, as its collateral (ETH, USDC) was depreciating but the over-collateralization ratio ensured DAI always had adequate backing.

UST (algorithmic type, no real collateral): Complete collapse. This is the ultimate lesson of stablecoins without real collateral — when market confidence collapses, no real asset can support it back to $1.

What this contrast shows: Collateral type is the core determinant of whether a stablecoin survives stress tests. Fiat-reserve and crypto over-collateralized types both passed the test during this crisis; algorithmic types with no real collateral went to zero.

Diagram
Stablecoin Collateral Types: Risk-Yield-Decentralization TriangleA triangle diagram with three axes: Safety (top), Yield Potential (bottom-right), Decentralization (bottom-left). USDC plotted near Safety apex, USDS in the midStablecoin Collateral Types: The Three-Way Trade-OffEvery stablecoin trades off Safety · Yield · Decentralization — no type wins all threeSAFETY(principal protection)DECENTRALIZATION(no central control)YIELD POTENTIAL(built-in returns)USDCUSDT · PYUSDCash + T-billsUSDSDAI · crvUSDETH + RWAUSDesUSDe yieldDelta-neutralFiat-Reserve (USDC)✓ Highest safety✗ No built-in yield✗ Centralized (bank risk)Best for: payments, savingsCrypto Over-Collateral (USDS)✓ Decentralized + yield ~4%✗ Capital inefficient (150%+)✗ Liquidation risk in crashBest for: DeFi yieldSynthetic (USDe)✓ High yield (bull: 20%+)✗ Funding rate risk✗ CEX counterpartyBest for: bull market yieldStablecoin Bible · stablecoin-bible.com
Feel free to share. Please credit the source.
Common Misconceptions +
✕ Misconception 1
× Misconception 1: 'Crypto over-collateralized stablecoins are safer than fiat-reserve because collateral is transparent on-chain.' Not entirely correct. Fiat-reserve type is actually generally more stable under normal market conditions, because its collateral (dollars) isn't affected by crypto market volatility. Crypto over-collateralized types can suffer bad debt during extreme market crashes (like 2020's Black Thursday) due to liquidation failure. High transparency ≠ highest safety; safety depends on which market stress scenario you're considering.
✕ Misconception 2
× Misconception 2: 'Synthetic stablecoins like USDe are fake dollars and definitely more dangerous than fiat-reserve.' Synthetic types aren't 'fake dollars' — they're a different mechanism for maintaining dollar value via derivatives hedging. In positive funding rate bull market environments, USDe's principal stability is comparable to USDC (main risk is exchange counterparty, not direct principal disappearance). Its dangerous scenarios are primarily prolonged negative funding rates and extreme market liquidity crises — not everyday conditions. Risk assessment for any stablecoin should be based on specific scenarios, not intuitive type judgments.
The Missing Link +
Direct Impact

Capital Efficiency vs Safety: The Core Trade-Off of Three Types

The three collateral types represent three different capital efficiency and safety trade-offs:

Fiat-reserve type has highest capital efficiency — $1 of dollars backs $1 of stablecoins, no waste. But safety highly depends on centralized institutions (issuer + banks), losing decentralization advantages in crypto-native contexts.

Crypto over-collateralized type has lowest capital efficiency — you need to lock $1.50 to borrow $1, with $0.50 in capital 'sleeping.' But it trades this for true decentralization (no centralized institution can control it) and better transparency (all collateral queryable on-chain).

Synthetic derivatives type tries to break through this dilemma — theoretically approaching 1:1 capital efficiency while not depending on centralized dollars. But the cost is the most complex mechanism risks: negative funding rates, exchange counterparty risk, insurance fund insufficiency.

Practical meaning for you: No type is optimal in all dimensions. The right approach is choosing by use case: high safety requirements (payroll, large value preservation) → fiat-reserve; decentralization requirements (don't want to depend on any institution) → crypto over-collateralized; seeking yield and understanding mechanism risks → synthetic. Holding a mix, letting different types' advantages complement each other, is the most rational strategy.

Ask a Question
Please enter at least 10 characters