The road to 1B FEI

PCV Development
1

As I shared in the Resume TRIBE Buybacks thread, I believe Tribe should focus more effort of making FEI a great product and increase adoption.

Indeed, a cool narrative would be to have each Tribe being fully backed by stacked ETH + the yield of additional investments. You can see my work on Asset Liability Management for stablecoin issuer here. Most of those idea are also discussed in my article Crypto Banking 101 (published almost one year ago).

Part of achieving this magic place is to get a lot of leverage from the stablecoin issued, i.e. FEI. A stablecoin is usually a quite stable source of funding and a cheap one at scale (MakerDAO pays only 1bps on DAI and there is too much demand).

I see two “simple” actions to provide FEI users a better value:

  • Be first class at keeping the peg
  • Provide a native risk-free interest rate

Happy to hear more thoughts about how to make FEI more competitive and used (one idea is to provide payment rails for web3 merchants but that’s way more complicated).

Be first class at keeping the peg

It is no surprise that FEI being a stablecoin it is pegged on the US Dollar. Yet the question how good the peg should be. I made a Twitter poll, and the anser is 22% for 0bps and 63% for 10bps (the rest for bigger spread).

Recent history shows that FEI was quite good at keeping the peg after the changes of FIP-72. Yet those last few days, it’s below the 10bps bar. Part of that is the 1bps FEI-USDC pool that is greatly incentivized but fully drained now.

image
image892Ă—388 24.2 KB

One corrective action would be to decrease the fees for the PSM from 10bps to 3bps. I use 3bps and not 5bps (range of the USDC-FEI pool) because there should be a buffer to execute trades (the PSM is based on DAI, but that can be swapped to USDC without cost). I would personally even go to 0bps as to be on par with DAI and FRAX (FRAX is not that hard pegged but have insane liquidity in Curve with an insane A factor so the result is a very good peg). It’s not like this PSM is providing much revenues (I guess around $20k for the last 2 months).

Provide a native risk-free interest rate

The whole idea of a stablecoin issuer is to issue low cost liabilities (FEI) and put that to work to generate revenues. In a competitive market, there is a need for providing value to your users. I understand you get 15% with governance reward by providing GUNI LPs on Fuse. While it’s good for people in this forum, it’s not good for real users. A simple stacking contract for FEI would provide a simple interaction for other smart contracts (Compound is integrated with MakerDAO DSR for instance) and even for exchanges (CeDeFi). The idea was already mentioned here.

Even a simple 2% would provide a great step in this direction as it would decrease the spread between the supply and borrow rate on money markets for instance (and integration of this feature on Fuse would be great putting FEI as the best stablecoin there to be deposited). in fact, this solution would be great for both FEI and Fuse (as other protocols will take more time to implement it).

To grow, you simply allocate for of your profits to FEI holders (not arguing about going as wild as Anchor).

What do you think?

10 Likes
2

good ideas

lowering the FEI-DAI PSM fees seems like a good move. I would want to see how the system behaves at 3-5 bps before going all the way to zero, but zero seems like the natural target to aim for. ideally FEI’s peg should have a tight spread and be above peg just as often as below, and for a PSM-based peg mechanism, this requires zero-ish fees or continual growth/revenue

Im also in favor of a experimenting with a savings rate. the tokenomics of a savings rate are potentially very good, especially if the PCV is earning a higher yield than the savings rate. I dont think the currently existing farms are too complicated for “real users” though…this is a UX issue and theres nothing preventing these farms could have a nearly identical UX to a savings rate staking contract

sidenote I would pushback on the framing of a “Great Stablecoin”. there are multiple dimensions for evaluating the quality of a stablecoin. pegs are the defining feature of stablecoins and they are easy to precisely measure so it’s natural to focus on them. however LUSD is a Great Stablecoin despite having a less precise peg. it’s great in different ways than other stablecoins that have ultra tight pegs

in terms of expanding FEI to 1B and beyond, I think the only way to do it is by offering unique products and services that generate FEI demand. we simply arent going to beat DAI or anyone else at their own game, instead we need to create a new game. keeping up with the competitive landscape of other stablecoins (and the tightnesses of their pegs) is obviously important, however I don’t think that having even a 0 bps peg would meaningfully distinguish FEI from the competition at this point

7 Likes
3

We might differ on the definition of real user :slight_smile: The farm imply to take a Gelato risk, Circle or DAI risk, a Fuse risk and get TRIBE token not FEI. They are not one year old and I’m quite confident they will not last 1 year. You can check how much is still in legacy pools.

I used the wording Great Stablecoin to let everyone decide what it was. LUSD is great in the sense of decentralization resistance but it might remain niche. My view of greatness is $1T PCV in 10 years.

I agree that a strong peg is not enough to succeed, it’s merely a requirement to survive and I welcome all ideas.

Speaking of which, with a native risk-free interest rate, we could be more competitive on money markets (as there will be no idle FEI earning nothing in a smart contract) but the same can be applied in exchanges. Either in direct integration (which basically mean creating a Dex at this stage and getting it integrated in aggregators) or through using ibFEI (interest bearing) as key item with unwrapping being done in the UI (1inch does that) or with a protocol owned ibFEI-FEI 1bps pool.

This is just following from the tagline Fei being the stablecoin of DeFi, okay so you need to be better for Dex and Money Markets for which a frictionless peg and a native interest rate is just a killer feature to drain from other stablecoins.

Surely, plenty can be done in DAO2DAO interaction but I’m less following this part.

2 Likes
4

Very excited by the title of this conversation lol

Thanks once again for your thoughtful discussions @SebVenture. I’ll share some of my ideas below, some are similar :slight_smile:

I see 3 main challenges on the road to 1B FEI :

  • Defending the FEI peg
  • Make FEI desirable to use and hold
  • Sustainably handle large amounts of PCV (to back a large amount of FEI)

Defending the FEI peg

I think we should move away from PSMs (and pseudo-PSMs) as we use them today, change their spread to 0.5% - 1% on redeem, 0% on minting, and create new mechanisms that make it desirable for users to market-make the FEI peg.

The protocol spends an immense amount in LM rewards to create “pseudo PSMs” (Uni-v3 FEI-DAI and FEI-USDC, Curve FEI-3crv) which create FEI demand and stability, but only initially. After these pseudo PSMs are exhausted, they actually act against the protocol, because a huge demand of FEI needs to be created before these LP positions soak enough FEI to go back to 50/50 proportions (= FEI at peg). The assets paired with FEI are also not in PCV, so the protocol don’t earn yield on them.

The protocol’s PSMs are an important guarantee to offer, but should be seen more as a last resort mechanism than a primary stability mechanism for people that have a high time preference, IMO.

I think the 0.99$ peg should be secured by PSMs, but the remaining 0.01$ should be backed by delayed perfect peg redeemability and bonds that create organic demand for FEI and marketmaking of the FEI peg.

Example of how the bonds / delayed perfect redeemability would work :

  • configure FEI bond to 3% APR and 1 week bond time (for instance)
  • user can stake 100k FEI on a contract
  • at anytime, user can unstake 100k FEI + DAI bond yield
  • if user stakes more than 1 week, they can claim 100k DAI instead of 100k FEI. They also still earn the DAI bond yield (e.g. 60 DAI for 100k staked 1 week at 3% APR)
  • because of gas costs, it’s going to be better for users to withdraw everything in the same token & wait 1 week, if they want to arbitrage the FEI peg.
  • some users might just use this as a “savings account”, essentially locking FEI out of circulation in exchange of a 3% DAI APR yield. Additional TRIBE LM rewards can be stacked on top of this bond yield.
  • it’s straightforward for yield aggregator protocols to create a FEI strategy that stake in this contract, collect DAI yield, and sell DAI for more FEI.
  • bonds could possibly be opened to assets not in the PCV (USDC, USDT…) because the protocol don’t hold/own them, they could just be market-bought assets which the redeemers/bonders find desirable & will eventually own

This would attract users to buy FEI when it’s below peg, because for instance if FEI trades at 0.995, they can spend 100,000 DAI to buy 100,500 FEI and then bond this FEI to get 100,560 DAI after 1 week, that’s an APR of 33% for DAI (part of the yield is the FEI bond yield, part of the yield is market-making FEI’s peg).

FEI bond yield can be set aggressively at first to increase PCV.

It’s more optimal for the protocol to incentivize the bonds with TRIBE (in addition to DAI yield or whatever the bond asset is), because users that buy FEI to bond will increase the PCV through the PSMs, and the protocol can market-make itself the FEI/DAI pair for instance, and earn trading fees, instead of renting this liquidity / “pseudo PSM” through LM rewards.

This creates obvious yield strategies for all assets supported by FEI bonds (user stake DAI, vault buys FEI with DAI, vault bonds FEI to DAI, vault ends up with more DAI than initially), which will increase the amount of liquidity in the ecosystem actively defending FEI’s peg… There could be official Tribe DAO yield products (Vaults?) that exploit this strategy.

Meanwhile, the FEI staked in the bonds engine is backed by PCV, which earns yield, so the cost of this 3% bond yield for users is partially/totally covered for the protocol (if CR is 200%, PCV earnings of 1.5%/year is sufficient to cover bond yields), not to mention an increased amount of FEI adoption & higher PCV means the protocol could possibly mint more FEI on its own to earn yield, too.

These bond yields make FEI delayed redeeming interesting even when FEI is at peg, so we can expect these bonds to be used all the time.

When FEI is above peg because too many people buy it to bond, there is a direct PCV inflow through PSMs that have 0 minting fee.

It’s like a flywheel to increase PCV using current & future PCV yield, and I think somewhat similar to Seb’s suggested risk-free interest rate (the FEI staked in the bonding contract is not re-hypothecated anywhere, in fact it could even be burned, and it earns a risk-free bond yield denominated in another token).

Make FEI desirable to use / hold

I don’t have much to add here. Having a stable peg, and have FEI at the center of kick-ass DeFi products sound like great ways to make FEI desirable to use.

Offering sustainable USD yield by holding and staking FEI somewhere sound like something that would make FEI desirable to hold (like a liquid savings account).

Sustainably manage CR at large amounts of PCV

To back a large amount of circulating FEI, a large amount of PCV is needed.

Large amounts of PCV exposes the protocol to a lot of risks (illiquidity of markets, volatility of assets, percent of leverage of circulating FEI vs PCV).

There is no liquidation mechanism or anything that protects the PCV from volatility, and it’s probably not desirable to have FEI 90% backed by other stablecoins (and it couldn’t scale above competitors), so at some point I think the Tribe DAO will need an in-house options protocol or leverage product, so that it can accept ETH collateral to mint FEI & sell the ETH volatility risk to someone else. I see this as a mandatory step on the road to 1B (user-owned) FEI, or at least on the road to 10B FEI.

3 Likes
5

I have been with Tribe since genesis, experienced Tribe-Rari merger.
As a wanna-be real user, I have no idea about what to do with Fei except borrowing/leveraging/staking.

I cannot use Fei to purchase any service/products or transfer but only play with ever-getting “complex” inward financial transactions(pools etc) which are trivial financial instruments in real world.

I am sorry but sitting on PCV 2.5 x circulation, Fei lost focus.

2 Likes
6

This is “the last mile” problem for the whole DeFi, i.e. DeFi<–>real world interface. Actually I’d argue that this is a major bottleneck containing DeFi’s and Fei’s in particular exponential growth.

First, being able to wire funds bypassing banks at a fraction of what they’d charge you isn’t small. The bigger your transaction, the bigger your gain. Imagine a company, which makes tens of thousands of transactions worth billions per year. Or a financial institution making millions of transactions worth trillions. Their gain just from using DeFi would be many millions. Plus a DeFi txn takes a couple of minutes vs a couple of days at a bank. But of course, if either of the parties isn’t DeFi-native, you’d have to find the solution to the on/off-ramp problem. Afaik, this problem is being tackled by some people within DeFi, would be good if Tribe joined these efforts.

Second, decentralised permissionless credit market isn’t small either. We’re witnessing in real time balkanisation of TradFi markets and DeFi might be a solution to this problem.

Also APY on Fuse is currently 12%, go find a TradFi bank with such deposit interest rate in USD.

Yes, other stables can also do that, but here it’s already a debate about algorithmic vs non-algorithmic stables.

7

The one and only KPI is “ FEI in circulation” and it is stagnant.

1 Like
8

This proposal is a tremendous opportunity to actualize the value proposition of FEI and create strong narratives that broader DeFi can see and touch.

Having a tight peg <5bps is very important for institutional and whale adoption, which is where the power law of benefits kicks in. I am very in favor of lowering the spread on the PSM.

The critical insight is that the PSM is just a “demand buffer” for users who wish to trade in and out of other extremely lindy and tight pegged stables like USDC and DAI. FEI would only need to hold enough DAI to back these large redemptions. This is “piggybacking” on the DAI peg.

By creating a native FEI interest rate analogous to the “DAI Savings Rate”, FEI can attract a true sink of deposits and grow its balance sheet to accumulate more stables and critically, ETH.

The most powerful piece of this narrative is that the Tribe DAO can place a flag in the ground that it will be preferentially accumulating ETH with protocol yield. ETH is even more potent than BTC which is now backing UST, due to native staking yield and deflationary properties.

Imagine if the Tribe DAO made an explicit goal of acquiring 1 million ETH

Every month, another 10k, every year another 100k. By 2030, the Tribe DAO would become the single largest ETH holding entity.

This is obviously subject to risk management necessary to back FEI and Tribe holder preferences.

11 Likes
9

I strongly agree with this strategy, however I think it’s critical that we offer long-dated bonds instead of a liquid savings rate. By locking FEI away for a long period of time, we can build the resilience necessary to survive a bear market and continue acquiring ETH without draining the protocol on a downturn.

5 Likes
10

I have a proposal of a ClearingDAO that would remove the need to hold DAI completely. By having bilateral agreement between stablecoin issuers, users will be able to swap one stablecoin to the other frictionlessly. It is also a way to migrate unused capital (e.g. MakerDAO) to protocol that are better at investing such capital and lack stablecoin demand.

I think it’s a key differentiation of Fei Protocol to hold a lot of ETH. I made some research for MakerDAO and it shows that holding quite some ETH is a good balance sheet decision for the current state of Tribe. There are strong limits to the leverage we can achieve but a strong point and a key differentiation.

If we achieve 1B user Fei, then the question will be what else beyond ETH?

We can start by giving 4% (the “ETH yield”) to Fei holders, transmuting USD to ETH without the risk.

5 Likes