By magnitude, the biggest question for the Defi ecosystem for the next cycle is who gets to keep the yield on treasuries backing reserve stablecoins such as USDC and USDT? If Secretary Bessent and the federal reserve are correct in their forecasts, we’ll be at something like $3T in stablecoin market cap by 2030. At even a modest fed interest rate of 3%, the interest on the assets backing these stablecoins will create about $90B in recurring annual revenue.
To put that in perspective Hyperliquid, which is touted as a monumental Defi success this year, is currently generating between $1B-2B in annualized revenue. PancakeSwap and Aave, the highest volume Dex the largest Defi money market respectively, make between $100M and $200M in annualized revenue. Even if secretary Bennett is off by an order of magnitude, the revenue from treasuries is likely to be bigger than all of Defi combined.
This $90B annualized revenue can either create a handful of outlandishly high value stocks at the direct cost of consumers or it can be distributed to make all of Defi flourish. Those are the stakes. Let’s dive in. First I’ll cover who are the possible beneficiaries of this revenue then I’ll write about how we can change the trajectory of our ecosystem from Tether dominant to something better.
There are four parties related to any reserve stablecoin:
- Issuer. These are the companies managing the reserves e.g Circle, Tether, or Paypal. Tether currently keeps 100% of the treasury yield on reserves backing USDT. If Tether is the answer to the big question, then they will be the most profitable company in the world per employee by far. For reference Tether has about 150 employees right now and aiming to IPO at a $500B valuation. For reference, Chase Bank with 315k employees, has a market cap of $800B. This is the most ridiculous outcome but is the one we’re headed towards if the banks get the changes they want into the market structure bill.
- Distributor. These companies are essentially the marketing arm of stablecoins e.g. Coinbase or Paxos. Coinbase has a profit sharing agreement with Circle. Coinbase receives all of the revenue for USDC on Coinbase and half of the revenue for USDC off-platform. This is the source of the yield Coinbase gives holders of USDC on their platform. An entity can fill multiple roles. For example, Tether is both an issuer and distributor.
- Application. In my opinion these are the most underappreciated and undercompensated parties in the ecosystem. A basic principal of justice is that people should be rewarded proportionally to their contribution. Within Defi most of the stablecoins are in some form of yield bearing position on an application e.g. Hyperliquid, Curve, Aave, etc. These applications are why the digital dollars are on chain. Without those applications the stablecoin market cap would be substantially lower. Even outside of Defi I think companies like Binance are undercompensated for the demand they create on USDT. If I stretch this argument I could apply the same reasoning to argue that L2s and L1s should receive something as well.
- Holder. All the stablecoins on chain ultimately fall under the control of some private key managed by a person or organization. If the holder isn’t receiving the treasury yield as an incentive to hold (like they do on Coinbase) then they are being robbed of it. They aren’t the only ones though; any yield-seeking stablecoin is being compensated by a different party for at least as much as the treasury yield. How this is paid can vary from the borrower paying higher interest rates to token inflation from the application paid to the LP but somehow or other the wider ecosystem is absorbing the time cost of money of every USDC and USDT on chain directly into Tether’s/Circle’s pockets.
Now, what can we do about this? Section 4(11) of the GENIUS Act prohibits stablecoin issuers from paying the holder of a stablecoin any form of interest or yield. IANAL but this doesn’t seem to prevent the distributor or application from receiving that yield. Even for the holder though, Coinbase is and has been paying users to hold USDC on their platform. They would argue they have consistently been a compliance-minded company, so how does that work? There are two forms of indirection at play. First, Circle (the issuer) pays Coinbase (the distributor) the yield. Circle does not pay the holder the yield, Coinbase does. Second, Coinbase doesn’t directly pay that yield to users for holding or transmitting USDC. Instead, the revenue from Coinbase is put into a rewards program. This is not dissimilar to how Visa takes fees from the merchant and pays it to credit cards users as points redeemable for cash. One would assume Coinbase would argue in court that Circle is compliant with the law because they are paying the distributor and nothing about this provision prevents the distributor or an application from creating a rewards program. Obviously none of this has been adjudicated, but it doesn’t look like the current “project crypto” SEC is going to pursue a case here at least. If this works and there isn’t something carefully crafted into the market structure bill to prevent this then this pattern can serve as a template for the entire space to ripple yield down the waterfall.
Let’s assume for the sake of argument that this structure is legal and is protected by Circle/Coinbase lobbyists or even just senators that don’t see how consumers are protected by being denied the yield on their own money. How can Defi use this to fund our applications and benefit users? We follow a Defi variation of the Circle/Coinbase pattern.
First, a pro-Defi issuer such as Paxos, Frax, or Stripe through Bridge make a stablecoin. This could be an ecosystem specific stablecoin such as USDH that pays out all the reserve yield to the application or it could be an ecosystem stablecoin such as Frax that bypasses applications and instead pays LPs on various applications. Next, that issuer uses the treasury yield to make more of that stablecoin for distribution (I assume they’ll keep at least a share of the yield as profit). Then that issuer sends the new tokens either to various application teams which internally take care of distributing the profits to holders or they do what Frax has been doing for years and distribute the new tokens as bribes. For bonus legal obfuscation application could pay holders in inflation rewards of their app token and using the stablecoin rewards for buybacks. Economically this is similar to mint/burn models such as EIP-1559. Legally speaking, the goal here is to make it easy to argue in court that the payments to apps/holders are valid rewards programs or incentives for providing a service on behalf of the distributor. Economically speaking, the goal is to more equitably distribute the treasury yield to better treat consumers and to incentivize stablecoin adoption.
With the distribution mechanism understood, we need a mechanism to bootstrap adoption and force an ecosystem migration off of the status quo stablecoins. Without that, arbitrage eats the first mover alive. I wrote about this in the first-mover disadvantage section of my Moloch’s Toolbox post. TLDR: you either need to make it economically profitable for first movers to absorb that cost or you need to enable people to coordinate a move without cost. I’m going to highlight two strategies we’ve already seen in Defi that can achieve this.
- Atomic Migration. We copy the bootstrap mechanism from Sushiswap. You create an escrow contract that LPs of the old pool deposit to which can migrate all of the assets in the contract at the same time subject to some prearranged condition. Smart contracts are beautiful for this. They can do so transparently and without middleman trust. This strategy was extraordinarily effective when paired with incentives but it can work as a standalone coordination mechanism. The main difference in this scenario versus Sushiswap is in this scenario we have to change the underlying LP from something like ETH/USDT to ETH/FRAX. To migrate atomically this smart contract will need to be able to swap potentially billions of the escrowed stablecoins to the new one with acceptable slippage. Since the redemption mechanisms of reserve stablecoins are never on-chain this can’t be done with a flashloan. It will probably have to be financed at least briefly by the vampire attacker e.g. borrow a few billion FRAX, honor an OTC 1:1 swap in the escrow contract, rebuild all the LPs with FRAX, then work on redeeming all the USDT post-migration. This is tricky and may cost tens of millions in borrow fees but it’s worth doing.
- Concentrated Emissions. If the problem with a migration is that a smaller pool faces higher costs, then one way to flip this problem on its head is make the new stablecoin the larger pool. Liquidity in Defi is provably responsive to rates; so through some combination of bribes and LDR votes that direct issuance at pools we concentrate as much power as we can as an ecosystem on a small number of pools to attract an enormous amount of liquidity for a short time. As liquidity migrates to the new position to lower its own costs, a long tail of the $90B annual treasury yield can be used to continue bribes as necessary to ensure that these pools remain the largest liquidity across exchanges and money markets in perpetuity. The needed magnitude of these bribes should drop over time as this new stablecoin becomes standard across Defi. Whatever it costs to finance this, there is a $90B a year carrot at the end of this path that I hope can make any costs worthwhile.
At the end of this, USDT and USDC won’t just go away but they will lose their dominance. USDC will remain prominent on Coinbase for the inherent yield is pays there and, if anything, we can expect more exchanges to follow this model. Those are basically application specific stablecoins where the application is the Cex. Binance will probably revive BUSD, Kucoin will probably create a KUSD, etc. The good news is each fragmentation of USDT liquidity weakens its’ network effect. Going from one ecosystem stablecoin to another will pass through Defi and use a Defi friendly stablecoin as the hub. By dominance, this Defi friendly stablecoin can overtake Tether. Then begins the long negotiation of where this yield goes to best reinforce this ecosystem. It can be used to fund a combination of infrastructure, gas subsidies via account abstraction, application development, and Defi LPs (holders). Even 1% of this revenue could fund the Ethereum Foundation in perpetuity.
On our current trajectory Tether gets all that yield and will be the most successful company per employee of all time. If we succeed at this all that value will trickle down to the market cap and end users of all Defi. This is a goal I think is worth organizing for.