The Anchor bailout

Do Kwon writes a $450m check to Anchor

Anchor’s native token, ANC, has rallied ~40% in the last 5 days, as Do Kwon announced a $450m grant from LFG.org (“Luna Foundation Guard”) to recapitalize Anchor’s depleted yield reserve.

This was pretty obviously incoming in my view — Matt Cantieri had made several loud references recently to bullish news incoming for the ANC token (they have quite a good track record of delivering on these quasi-promises), and Do Kwon had polled Twitter for an ANC bailout of $100m / $200m / $300m ($300m won easily). I had moved some money from ANC-UST LP pools to ANC governance tokens in anticipation.

I don’t know if the math is exactly the same, but I think of a token-stablecoin LP position as being 50% long the native token and 50% short the volatility of the native token, with the caveat that your native token exposure goes down as the native token goes up in stablecoin price, and vice versa. The ANC-UST LP position has not been fun, and my LP position had a drawdown of around 30% as ANC dropped 70% from its frothy December highs.

Anyway, the Anchor bailout is being funded by minting 9m LUNA tokens … and selling them in the open market.

As you can see from SmartStake’s Terra blockchain analytics, before this event, the lifeblood of LUNA — growing demand and issuance for UST — had pretty much stopped.

(The massive November spike marked the implementation of LUNA’s Columbus-5 update, in which UST surplus income was returned to LUNA holders via buying back LUNA and burning it — which increases the supply of UST.)

As MIM’s money supply contracted by 40%

¹

over the 4 days ending on February 1st, UST demand growth had flipped into negative territory. The Degenbox is a lead weight around Anchor’s ankles until Anchor can find new borrowers to increase their return on invested capital.

Anchor needs time to build out solutions to increase its leverage, by expanding on-platform collateral eligibility. The Degenbox and its $1B of money creation seemingly did not result in increased transaction velocity on the Terra network, causing the staking yield to stagnate (even when volatility and transaction count went up significantly). This resulted in a drop in the staking yield and an effective outflow of capital from the Terra ecosystem.

Anchor wants to remain a decentralized, open protocol, and allow the Degenbox to stick around. This is a noble goal, but will be costly, especially if Terra’s transaction velocity continues to stagnate.

However, the Degenbox wasn’t as big of a problem as I’d first thought. Since the Degenbox’s money went straight into LUNA staking yields, it wasn’t that bad of a deal for Anchor, unless LUNA’s staking yield gradually dropped.

This was the most detailed breakdown in the Discord that I saw. My current understanding of the model is similar:

My model comes with a big caveat, though. Anchor is sitting on a huge amount of UST. It can choose to take that UST, turn it into LUNA, and earn LUNA staking rewards at 8.2%, thereby nearly halving its net deposit cost.

However, I don’t think Terra wants Anchor to do that. I think Anchor’s UST sits idle in the event of a LUNA death spiral, in which case, ANC’s massive UST deposits could convert themselves to LUNA at a discount to peg, and stop a potential run on the Terra ecosystem. So Anchor instead has to idle the deposits and eat losses.

As a result, Anchor is currently burning around $1.8mm of cash per day ($657m/year), not counting ANC emission incentives to borrowers. Its main expense is paying depositors a 19.5% yield on their deposits. Its main revenue source is bonded assets (bLUNA, bETH, soon to be bATOM and bAVAX I think) on protocols, getting a staking yield on them, and also incentivizing people to borrow against bonded collateral.

The bonded-asset part of the model always worked well, and was profitable. It was an innovative form of leverage, heavily subsidized in ANC emissions.

But the deposit side of the equation raced far ahead of the bAssets side of the equation after the Degenbox arrived. The extent of the cash burn it caused was initially hidden because, as the $1.1B of Degenbox UST pumped up the price of LUNA, it also pumped up the value of Anchor’s bLUNA deposits. But once the crypto-crash happened in 2h January, Anchor was saddled with most of the Degenbox chronic outflows, while the value of its bAssets had dropped. So the cash burn accelerated.

Anchor needs to find more ways to bond more assets and extend leverage against those bonded assets. The $450m from LFG gives Anchor another 9 months or so to get that job done. Anchor could’ve also “cheated” by putting more UST back into LUNA and staking it, but I’m guessing that’s a red line that Do Kwon didn’t want crossed.

1. https://dune.xyz/queries/212280 (click “line chart” to visualize).