Update OUSD Metapool amplification parameter

[micah]

Summary:

Increase the amplification coefficient (A factor) on the OUSD Metapool (factory-v2-9) from 100 to 500.

Abstract:

Origin Dollar (OUSD) is the self-custodial, yield-generating stablecoin. It is fully collateralized at all times and automatically distributes DeFi yields in the form of increased token supply. The price of OUSD remains pegged at ~$1.00 but the balance of each account holding the token increases at least once per day. OUSD can be redeemed by anyone at any time for a mixture of DAI, USDC, and USDT. The protocol has successfully scaled up to $300 million and back down to the current circulating supply of $51 million without any issues. OUSD is traded on KuCoin, Gate and Uniswap in addition to most of the volume occurring on Curve.

Motivation:

According to the Curve documentation:

The amplification co-efficient (“A”) determines a pool’s tolerance for imbalance between the assets within it. A higher value means that trades will incur slippage sooner as the assets within the pool become imbalanced.

Note: Within the pools, A is in fact implemented as 1 / A and therefore a higher value implies that the pool will be more tolerant to slippage when imbalanced.

The docs go on to suggest that the pools with the highest A factors should consist of “redeemable assets” as opposed to pools requiring lower A factors consisting of “uncollateralized algorithmic stablecoins” and “non-redeemable, collateralized assets”.

OUSD not only supports permissionless redemptions, it is collateralized by the same stablecoins as 3Crv, which is the other side of the pool in question.

The OUSD Metapool’s current A factor is considerably lower than all of the other top 10 USD-based metapools by TVL:

Pool		A
OUSD/3Crv		100
3pool		2000
Frax/3crv		1500
Frax/USDC		1500
MIM/3Crv		2000
alUSD/3Crv		200
LUSD/3Crv		500
sUSD/3Crv		256
Compound		4500
TUSD/3Crv		600
USDD/3Crv		200

The A factor for the OUSD Metapool was last updated in September of 2021 with an increase from 10 to 100. At the time, OUSD’s circulating supply was around $20 million and daily trading volume rarely exceeded $50k. Since then, the daily trading volume has exceeded $10 million on multiple occasions. An increase to the A factor would facilitate larger trades with less slippage and continue to generate greater fees for Curve LPs and CRV stakers.

We are proposing to increase the A factor on the OUSD Metapool over a one-week period with a recommendation to go from 100 to 500 at this time.

For:

• Better peg stability for an asset fully collateralized by the same tokens as 3pool
• Reduced slippage to support large trades
• Increased trading volume and additional fees

Against:

• Low existing trading volume compared to some other pools

[chanho]

Nicely written. Might be the rare time I see a legit reason in the “against” list put by the proposer!

[nagaking]

Our analysis suggests that this is a reasonable parameter change. The current A is likely too low, and raising to A=500 should reduce slippage and increase returns for this pool.

As the proposal notes, OUSD is backed by the same tokens as 3CRV, and therefore should be expected to trade somewhat in line with 3CRV. Consistent with this, its price has remained mostly between .99 and 1.01 since June 2021. While the value of OUSD is not always equal to 3CRV due to fluctuating asset balances and differing redemption fees/mechanisms, the two are still expected to trade somewhat in line with one another due to similar, largely stable collateral.

Speculatively, raising A might also stabilize OUSD price somewhat, given that it is primarily traded on Curve and Uniswap, and occasionally sees greater volatility than would be expected given its backing collateral.

Source: Origin Dollar Price, OUSD Live Chart, All-Time High & Market Cap (USD) | Nomics - Note that there is more extreme volatility prior to June 2021, but this may be due to bad data or early “growing pains”.

Simulations
Initial simulations produced quite noisy results, which can occur for tokens with less volume. We therefore (a) fit a quadratic function to “smooth” the annualized returns, and (b) simulated the pool allowing for up to 10x the observed volume (see below). Both of these analyses suggested that A=500 is within the range of reasonable values.

Balance of Reserves

Using the observed volume, the median balance was .70, corresponding to ~65%/35% OUSD/3CRV balance. With 10x volume, the median balance was .68. While these values are slightly below our typical target of .8 (60/40 balance), this concern is largely mitigated by the risk-mitigating factors discussed above.

Returns
The quadratic fit used to “smooth” the annualized returns peaked at A=431. In the simulation using 10x the typical volume limit, returns appear to approach their saturation point near A~500. Therefore, these results also point to an A value in the 400-500 range. This may suggest that an A closer to 400-450 might be preferred but, again, the low-risk profile of OUSD mitigates our concerns here.

---

Initial Simulation:

Quadratic Fit to Annualized Returns
For the model fit, peak annualized returns are at A=430.79

Simulation with 10x Volume Limit

[micah]

Thank you for the thorough analysis.