Mountain Protocol - USDM Questions and Ideas
February 21, 2025
☼ Crypto
☼ Payments
This post presents questions and thoughts I have after reviewing the Mountain Protocol Documentation. Additionally, I explore potential structural changes to USDM, big-picture implications of stablecoins, and future directions for Mountain Protocol.
30-Second TL;DR
- Should USDM’s peg fluctuate based on yield rewards instead of staying at 1:1? Could this simplify rebasing, change how users time their investments, and improve the coin’s utility?
- Are stablecoins really decentralized, or is everything still built on Traditional Finance? Even algorithmic stablecoins rely on fiat-based stablecoins - so where does DeFi really begin?
- What happens if the S&P 500 launches a stablecoin? Wouldn’t it be classified as a security? Aren’t all yield-bearing stablecoins securities? Did Gary Gensler have a point?
- Is Circle’s USDC model extractive? While inflation hurts holders, it benefits Circle’s owners through superior yield - does USDM present a fairer alternative?
- What’s next for USDM? Can it expand through PSMs, LPs, and fiat on/off ramps to become a superior base stablecoin?
- How should USDM reserves be stress-tested? Could a risk-hedging agent profit when stablecoins face crises like USDC’s SVB depeg?
Table of Contents
- Yield & Market Dynamics
- Stablecoin Classifications & Systemic Risk
- USDM’s Future & Competitive Positioning
- Reserve Management & Stability
Yield & Market Dynamics
Should USDM’s peg fluctuate with yield rewards? Would this affect user behavior, simplify rebasing, and improve integrations in LPs & PSMs?
Should the Peg Fluctuate Instead of Minting New Tokens?
Right now, Mountain follows: balanceOf(account)USDM = shares(account) * rewardMultiplier
This means:balanceOf(account)USD = balanceOf(account)USDM * (USD/USDM peg), where USD/USDM peg is always 1:1 on Mountain’s platform and held through arbitrage on secondary markets.
But what if the peg fluctuates based on the rewardMultiplier?
If we redefine the equation as USD/USDM peg = rewardMultiplier,
Then, balanceOf(account)USD = balanceOf(account)USDM_Proposed * (USD/USDM peg).
Ensuring balanceOf(account)USDM_Proposed = shares[account].
- Right now, USDM uses a 1:1 peg, with rewards distributed separately.
- Could rewards be baked into the peg itself, eliminating the need to mint new tokens?
- Would this change how users time their purchases—like wanting to buy into the S&P 500 earlier rather than later?
- Would it make rebasing less technically complex, especially for wUSDM?
Liquidity & Arbitrage Considerations
- Would a fluctuating peg make USDM easier to integrate into liquidity pools (LPs) and peg stability mechanisms (PSMs)?
- If this system is too complex, could another linked token be created where the price fluctuates instead of USDM?
Low-Yield Environments
- If Treasury yields decline, how does USDM remain attractive?
- Does Mountain expect circulation to fluctuate with yield levels, or are alternative strategies in place?
When is “Day 0” for USDM?
- When exactly does USDM’s lifecycle begin—is it the day it’s minted, transferred, or reset with each transaction?
Stablecoin Classifications & Systemic Risk
Are stablecoins actually decentralized, or is everything still dependent on TradFi? If major financial indices launched a stablecoin, would it be classified as a security?
What if the S&P 500 Launched a Stablecoin?
- Would it be classified as a security?
- How does this compare to USDM’s classification?
Is DeFi Just Traditional Finance in Disguise?
- Even algorithmic stablecoins ultimately rely on collateralized stablecoins.
- Are all stablecoins just TradFi wrapped in a DeFi interface?
- If governments still back everything, but markets make the decisions, is this the real definition of DeFi?
- Could current stablecoins unlock efficiency gains?
USDM’s Future & Competitive Positioning
Can USDM replace USDC as a fairer stablecoin? What integrations (PSMs, LPs, fiat rails) will drive its long-term adoption?
- USDM is not extractive, unlike USDC, where Circle’s owners pocket all the yield.
- Circle profits from inflation—as interest rates rise, its yield grows, while USDC holders lose purchasing power.
- Wouldn’t it be better if USDM replaced USDC as the default stablecoin?
What’s Next for USDM?
- Mountain seems focused on:
- Swap pools
- Lending market integrations
- PSMs for stablecoin swaps
- LP adoption for yield farming
- PSMs and LPs are likely the core use cases, while swaps & lending require higher volumes than USDM currently has.
- Expanding native USDM on multiple blockchains could accelerate adoption.
Fiat Rails & Diversification Beyond USDC
- Can USDM add fiat on/off ramps to convert fiat directly into USDM? It might serve as a good use case for cross-border-payments
- Should USDM diversify beyond USDC, given its single point of failure risk?
- Would this make USDM a more sustainable base stablecoin?
Reserve Management & Stability
How does Mountain Protocol manage liquidity and risk? Can stress-testing stablecoins unlock new profit opportunities?
Liquidity & Maturity Management
- What processes and automations does Mountain Protocol use to manage the weighted average maturity (WAM) and weighted average life (WAL) of USDM reserves?
- What does the USDM minting-to-reserves workflow look like?
Reserves Attestation & Stress-Testing Risks
- Could a new risk-hedging agent profit by arbitraging stablecoin depegs?
- For example, if USDC loses its peg due to a banking crisis, could this agent trade around such events?
- Can this agent detect interlinked risks, like USDM holding USDC, which in turn may rely on the same banks as USDM’s reserves?
- Could DebtCoin hedge against U.S. debt servicing risks in the unlikely case of government defaults?
Final Thoughts
Stablecoins are evolving, and USDM is proof that there’s plenty of room for innovation beyond USDC and USDT!