The stablecoin duopoly is ending

Three's a crowd

Circle equity is worth $30.5b. Tether’s topco is reportedly raising capital at $500b. Together, the two stablecoins boast $245b in supply, or about 85 percent of the stablecoin market. Since inception, only Tether and Circle have maintained meaningful market share. No one else has even come close. Dai peaked at $10b in early 2022. Terra’s doomed UST did surge to $18b in May 2022 but accounted for only about 10% of the market, and then only ephemerally. The most ambitious attempt to dethrone Tether/Circle was Binance’s BUSD, which peaked at $23b in late 2022 for 15% of the market, before being effectively shuttered by NYDFS.

Stablecoin relative supply share. Data: Artemis

The absolute lowest market share print for Tether/Circle I could find¹ was 77.71% in December 2021 when Binance USD, DAI, FRAX and PAX combined for a meaningful share. (I suppose if you go back to pre-Tether, there was a time when it had no market share, but no major stablecoin from the pre-Tether era (Bitshares and Nubits) made it.)

The dominance of the big two peaked in March 2024 at 91.6% overall of stablecoin supply but has been slipping since then. (I am computing market share in supply terms since that’s easy to compute, but if you were to do it by transaction value, number of pairs traded against, real world payments, active addresses, or real-world payment volume, it would undoubtedly be bigger.) It has however fallen to 86% since its peak last year and I believe it will keep falling. The reasons are new assertiveness by intermediaries, a race to the bottom with yield, and new regulatory dynamics post-GENIUS.

Intermediaries are rolling their own

In the last few years, if you wanted to issue a white-labeled stablecoin, you had to be able to afford very high fixed costs, and you had to call Paxos. That’s changed. Today, you can choose from Anchorage, Brale, M0, Agora, or Bridge (Stripe). We have in our portfolio tiny seed stage startups that have successfully launched their own stables with Bridge. You don’t need to be a behemoth to do it. In his launch post about Open Issuance, Bridge cofounder Zach Abrams explains why rolling your own makes sense:

For instance, if you’re using an off-the-shelf stablecoin to build a neobank:

a) You can’t fully access the rewards to build the best possible savings account,

b) Your reserve mix cannot be customized to support increased liquidity or higher earnings, and

c) You’d have to pay a 10bps redemption fee to withdraw your own money!

He’s right. If you use Tether, you’re probably not getting any yield to pass along to your clients (who, at this point, expect some yield in return for taking their deposits). If you use USDC, you might get yield, but it’s a negotiation, and Circle has to take their cut. You have no control over your own destiny with regards to freeze/seize policy. You lack configurability. You can’t decide what blockchain to launch on. And you have to pay redemption fees (which could rise at any moment).

I used to think network effects would win out, and we’d end up with just one or two stables, but I don’t believe that anymore. Cross chain swaps are getting more efficient by the day, as is intra-chain inter-stablecoin swapping. In a year or two, I think many intermediaries in crypto will just show your deposits as generic “dollars” or “dollar tokens” (rather than USDC or USDT) and they’ll guarantee you redeemability in a stablecoin of your choice.

Already we are seeing this with a lot of the fintechs and neobanks that are product rather than crypto first. They care most about UX and would rather offer clients the best experience instead of honoring crypto traditions. So they just show your balance in USD and manage the reserves on the back end.

And intermediaries – whether they are exchanges, fintechs, wallets, or DeFi protocols – have a very strong incentive to strip out the major stablecoins and direct user flows into their own. The reason for this is very simple. If you are a crypto exchange with $500m in USDT deposits, Tether is earning around $25m/year on that float, and you’re getting nothing. There’s three ways to turn that inert capital into a revenue driver. You could beg the issuer to share some of the underlying yield. Circle does this through reward programs (although, to my knowledge, Tether doesn’t pass on yield to intermediaries). You could partner up with one of the newer stablecoins like USDG, AUSD or Ethena’s USDe that offers a yield revshare by design, or you could create your own stablecoin and internalize all of the yield.

In the above example, as an exchange, you’d have to convince your users to desert USDT and adopt your newer stablecoin. An obvious tactic would be an “earn” program whereby you offer the T-bill rate and keep some margin, say 50 bps for yourself. If you are a fintech product that serves a non-crypto native clientele, you may not even have to do an earn program or similar inducement. You could simply display user balances in generic USD and then swap them into your own stablecoin. It’s simple enough to swap into Tether or USDC at the point of withdrawal if necessary.

We are already seeing this happen. The latter flow with fintechs is the default for startups we see today. Exchanges are eagerly striking revshare deals with stablecoins. Ethena has been particularly successful selling into exchanges with this strategy. Circle famously splits interest revenue with Coinbase, who in turn passes it along to exchange clients holding USDC balances. Other exchanges are banding together to create their own stablecoin consortia. Notably, the Global Dollar consortium consists of Paxos, Robinhood, Kraken, Anchorage, Galaxy, Bullish, and Nuvei, in addition to over a dozen notable other partners.

Importantly, DeFi protocols are now exploring their own stablecoins. They can’t as easily swap out user deposits into other stables, but they can gently encourage their users to use one stablecoin over other. The most notorious case here is Hyperliquid, which underwent a very public bidding process for its stablecoin, with the explicit goal to reduce dependence on USDC and capture reserve yield for the protocol. Hyperliquid collected bids from Native Markets, Paxos, Frax, Agora, Sky (Maker), Curve, and Ethena, eventually – and controversially – going with Native markets. Today there’s around $5.5b USDC on Hyperliquid, or 7.8% of overall USDC supply. Although Hyperliquid’s USDH won’t displace USDC overnight, this public process was an optics loss for USDC, as other DeFi protocols will look to do the same. And conversely, we are seeing stablecoins trying to create their own DeFi ecosystems. The assumption of baked-in yield just creates a gigantic design space that has yet to be explored. It’s like if ordinary fintech apps were built on top of money market funds; or brokerages just natively paid out your yield from securities lending.

And we are even seeing wallets roll their own stables. Phantom today announced Phantom Cash, a Bridge-issued stablecoin with embedded earn and debit card functionalities. Phantom cannot require that their clients use one stablecoin over another, but there are endless levers they can pull to get users onto their own Cash product.

Based on the declining fixed costs of issuing a stable (or simply partnering with an issuer that is willing to strike a revshare deal), it no longer makes sense as an intermediary to surrender the revenue from your float to a third party stablecoin. If you are large and credible enough for users to trust your white-labeled stablecoin, you might as well do it.

Yields are racing to the bottom (top)

If you look at the chart of stablecoin supply excluding USDT and USDC, you can see a kind of regime change in the “everything else bucket” in recent months. In 2022 there was an explosion of also-rans, led by Binance’s abortive BUSD, and Terra’s catastrophic UST (not pictured on the below chart). After the credit crunch took down Terra and many other firms besides, the industry reset and a new crop of stablecoins emerged from the wreckage.

Stablecoin supply excluding USDT and USDC, RWA.xyz

Today there’s more ex-Tether/Circle stablecoin supply than ever before, and distributed among a wider set of issuers. The top names in today’s regime are Sky (Maker/Dai’s latest iteration), Ethena’s USDe, Paypal’s PYUSD, and World Liberty’s USD1. I think it’s also worth paying attention to emerging names like Ondo’s USDY, Paxos’ USDG (issued as part of a consortium), and Agora’s AUSD. Many other new stablecoins – including bank-issued ones – will be entering the game soon, but I think the data is already telling. There are a lot more credible stablecoins than there were during the last surge, and they collectively have more supply than they did in the prior bull market – even against the backdrop of Tether and Circle continuing to dominate market share and liquidity.

There’s something interesting about a lot of these new issuers. Many of them are focused on passing along yield. Ethena’s USDe, which passes along the yield from crypto basis trade, is the biggest success story of the year, surging to a $14.7b supply. USDY (Ondo), SUSD (Maker), USDG (Paxos/others), and AUSD (Agora) are built with yield-sharing in mind. Now you might protest and say “GENIUS prohibits yield provision” and that’s true, in a sense. But if you’ve paid attention to the histrionics of the bank lobby lately, you can tell that the issue isn’t settled. GENIUS doesn’t actually prohibit third-party platforms or intermediaries from paying rewards to stablecoin holders (which are, in turn, paid to the intermediaries by the issuers). Mechanically, I don’t think you could even design language to close this “loophole.” Nor should it be.

As GENIUS loomed and then passed, I noticed a shift away from direct yield-bearing stablecoins (that pay holders directly – explicitly prohibited by GENIUS), to issuers that pay via intermediaries. Circle of course does this with Coinbase, and they show no signs of stopping. Virtually every new stablecoin has some yield strategy baked in. And it makes sense. If you want to persuade someone to drop the highly liquid and established Tether for your stablecoin, you’re going to have to give them a damn good reason to do it. This was a prediction I made in 2023 at Token2049, and even though GENIUS put a dent in my timeline, it’s clearly happening.

So the fact that stablecoins are generally moving towards yield provision hurts the incumbents who are less flexible. Tether doesn’t pay yield at all. Circle has a rewards program with Coinbase and unclear relationships elsewhere. I think newer startups will be able to undercut the major issuers on yield and create a race to the bottom (or realistically, the top) phenomenon. This actually could benefit those with scale, as we saw with the ETF expense ratio race to 0 and the emergence of the Vanguard/Blackrock duopoly. But will Circle and Tether be the winners in this scenario, if the banks are still on the sidelines waiting to join?

Banks can play ball now

Post-GENIUS, and following rule changes at the Fed and other major financial regulators, banks are now free to issue and engage with stablecoins, without requiring a new charter. Bank-issued stables under GENIUS still need to abide by the rules – 100% HQLA backing, 1:1 convertibility on demand, disclosure and audits, submitting to the relevant supervisory authority. Stablecoins would not be considered insured deposits and banks cannot lend against the backing collateral. Whether a bank would want to issue a stable is an interesting question. When banks ask me if they should do it, I normally tell them it’s not worth the trouble, and that they should simply incorporate existing stablecoins into their core banking infrastructure without issuing directly.

However, a bank or consortium of banks might consider issuing anyway. I am confident we will start to see these in the next couple years. Even if stablecoins are effectively a form of narrow banking and their issuance would actually de-lever the bank, there are still revenue opportunities that emerge from a vibrant stablecoin ecosystem (custody, transaction fees, redemption fees, API integrations). Banks might also panic if they see deposit flight happening – especially if stablecoins are able to offer yield via intermediaries – and determine that issuing a stablecoin of their own is the best way to stop this happening.

Stablecoins also aren’t structurally all that expensive to issue, as a bank; you don’t need to hold regulatory capital against them. They are full-reserve, off-balance sheet liabilities with a lower capital intensity than ordinary deposits. Some banks might decide that they want to get into the “tokenized money market fund” game, especially if Tether’s profitability holds up. If the stablecoin meta ends up being that absolutely no interest sharing can occur, and all the loopholes are slammed shut, issuers have a license to print money (collecting 4% and paying nothing to their clients – even better than the NIM on “high yield” savings). In reality, I expect that the yield “loopholes” will not be closed and issuer margin collapses over time. But even if issuers are only keeping 50 or 100 bps, the biggest banks marshal trillions of dollars of deposits, so this is still meaningful revenue, if a shift from deposits to stables occurs.

All of this is to say, I expect that banks will indeed join the stablecoin game as issuers, for one reason or another. Earlier this year, the WSJ reported that JPM, BoFA, Citi, and Wells have held early talks about creating a stablecoin consortium. I think a consortium makes by far the most sense, as no bank individually has the ability to create the necessary distribution for a stablecoin which could compete with Tether.

Closing thoughts

I used to adamantly believe that we would only need one or two major stablecoins, or maybe a half dozen at most. “Network effects and liquidity are king,” I would repeat, somewhat incuriously. But do stablecoins actually benefit from network effects? They’re not the same kind of business as Meta or X or Uber. It’s actually the blockchain that is the network, rather than the token. If you can swap in and out of the token frictionlessly, and swap between blockchains quickly and cheaply, the network effect starts to matter less. If exit costs trend to 0, you can’t force users to stick with you forever. What the majors do have – especially Tether – is tight spreads against major FX pairs on hundreds of exchanges globally. That’s harder to beat, but I’m seeing (and funding) an explosion in providers that are linking stables to local fiat both on- and off-exchange at wholesale FX rates. They aren’t as worried about which stable they’re trading, as long as it’s credible. GENIUS fixed a lot there. Infrastructure maturation makes life better for everyone, with the exception of the big incumbents.

A few pieces have come together to poke holes in the Tether/Circle duopoly. Cross chain swapping has become a lot better and cheaper. Same-chain swaps for stablecoins specifically has become almost free. There are now clearinghouses to facilitate stable-stable transactions regardless of originating and destination stablecoin or blockchain. GENIUS is homogenizing (US-domiciled) stablecoins so infrastructure providers are taking less risk by holding them on their balance sheet. We are experiencing a fungibilization of stablecoins. The incumbents do not benefit from this.

There is now a wide set of white-label issuers driving down the fixed costs of issuance. Nonzero Treasury yields create a strong incentive for intermediaries to monetize their own float and strip out Circle and Tether. This applies most to fintech-style wallets and neobanks that have no allegiance to crypto UX patterns, followed by exchanges and now DeFi protocols. Every intermediary is looking greedily at user float and wondering why they aren’t turning that into revenue.

The prospect of yield – stunted by GENIUS but not eliminated – also gives newer upstart stables new degrees of freedom to compete with USDT/C who are more reluctant to to share yield. If, as I expect, the yield “loophole” remains in place, there will be a race to the bottom as far as yield sharing is concerned and this could erode the incumbents’ market positioning if they don’t react fast.

And the elephants in the room are the sidelined financial institutions with literal trillions of balance sheet. They are wondering if stables will cause a deposit flight, and what they should do about it if so. GENIUS and finreg shifts allow banks to compete. If they do, the roughly $300b of stable capitalization will look like child’s play. Stablecoins are only 10 years old. The race has only just begun.

1. Both Artemis and RWA.xyz whose data I used for this article exclude Terra’s UST in their datasets, so in practice I think their market share actually dropped to 71.7% during Terra mania. But many would argue it wasn’t really a stablecoin