Thread by @samkazemian

Some thoughts on $FEI.. 🧵

(obviously take this with a huge grain of salt as $FRAX is the only other algo coin in the @Uniswap top10. So it goes without saying I have some bias)

But as someone deep in the algo stablecoin space, I think there's important predictions to make.

Firstly, I won't comment on the $1B+ genesis ICOish raise because this is a technical/mechanism overview. I don't like to comment on speculation and the drama side of crypto.

$FEI's liquidity collateralization is an innovative idea in the algo space and deserves props. Their "PCV bonding curve+reweights" are well branded and a coherent concept in onchain monetary policy. It's economically sound, and I give major props there.

In my opinion, the most innovative idea behind $FEI is that liquidity is just as important as collateralization/redemption and they solve this in a single step by essentially collateralizing $FEI with FEI-ETH Uniswap liquidity. They blur the line between liquidity and collateral.

We had the same idea with $FRAX early on by wanting to accept FRAX-USDC and FRAX-DAI LP tokens as collateral for minting FRAX but opted to keep the v1 simple. We currently employ a @CurveFinance AMO strategy by backing $FRAX with FRAX3CRV liquidity. docs.frax.finance/amo/curve

The other $FEI innovation of choice, direct incentives (DI), makes little sense in my opinion and is possibly even worse than v1 algo coin bonds of 2020. DI punishes sellers when the open market price of $FEI is below the peg and rewards buyers during this period.

Put in another way: As soon as the market price of $FEI goes below peg, the protocol itself REMOVES LIQUIDITY AT THE PEG. The protocol literally never honors $FEI at the peg price ever, unless it would have traded at that price on the open market anyway. *queue rug jokes*

Consider the current "actual" price of $FEI on Uniswap that you can sell for on the non-DI pairs. It's $.95 ish. But you can never get this price from the protocol's own liquidity because the protocol is supposedly trying to "punish sellers below the peg."

The REAL price of $FEI is actually much, much worse than the market price. Currently, there is a ~.28 $FEI penalty if you sell right now. Effectively, $FEI is ~$.67.

However, if you buy $FEI right now, you will receive more than 1 $FEI per $1 of ETH. $ETH is worth $2,040 right now so you would receive 2,182 $FEI. When $FEI returns to peg (which is hard to know when), you'll have made a nice profit.

The idea here is to bake incentives directly into the stablecoin itself so that speculation is softened out. But this is as misguided as v1 ESD coupons of last year. DI essentially rugs the protocol's own liquidity on holders and forces them to hold until recovery.

Ironically, this actually INCREASES volatility away from the peg rather than stabilizes prices around the peg. $FEI doubles as both a stable ish token AND a speculative asset/game of when you will get rugged by direct incentives.

We learned from $ESD v1 that the stablecoin itself being the object of both stabilization AND speculation on future growth of the system is an absolutely terrible idea. That's why $ESD v2 is correctly borrowing/improving on the $FRAX 2 token model which separates out primitives.

In DeFi, separation of primitives is important. It allows proper economic activity around them and brings the highest amount of efficiency. Never pack both speculative forces AND stability forces into 1 token. It's already hard enough to stabilize a coin onchain.

That said, direct incentives will work as often as ESD coupons/Basis bonds. Namely, sometimes they will, mostly they won't, and it will become unpredictably volatile. So I assume DI will work to return the peg at times and not other times which is when PCV reweights kick in.

Here are my $FEI predictions:
1.) DI will be one of the first things to be soft pivoted out in the next 6-8 weeks after enough data shows this indeed does not work as well as hoped.
2.) The next PCV curve $FEI will launch will likely be some stablecoin pair like DAI.

3.) $TRIBE will become the speculative primitive in the protocol in a more formal setting like an actual backstop mechanism or a TRIBE PCV curve essentially transitioning the entire thing into a 2 token, fractional-algorithmic protocol. Sound familiar? 🙃

tl;dr: $FEI PCV+reweight is very cool, and I am a big fan. DI actually increases volatility instead of decreases. The speculative component needs to be separated out of the stable token. $TRIBE will formally take that speculative role soon. $FEI will look more like $FRAX if so.