Skip to content
Work with us

DeFi's Distribution War Moved to the Stablecoin

Stablecoin distribution is becoming a source of recurring revenue in DeFi. Explore how Hyperliquid, Phantom, and others capture stablecoin reserve income.

Share
DeFi's Distribution War Moved to the Stablecoin
In this report · 14 sections

Overview

Key takeaways

  • Stablecoin distribution can generate recurring revenue when platforms retain user balances and secure a share of the underlying reserve income.

  • Hyperliquid, Phantom, MetaMask, Jupiter, and MegaETH demonstrate different approaches to capturing and allocating stablecoin reserve income.

  • The profitability of stablecoin distribution depends on eligible balances, interest rates, revenue-sharing terms, retention costs, and operating expenses.

On 14 May 2026, Coinbase announced that it would become USDC's treasury deployer on Hyperliquid, where USDC balances across the ecosystem stood at approximately $5bn, and that USDH, the ecosystem's alternative stablecoin, would be gradually wound down. Under Hyperliquid's AQAv2 framework, deployers must share approximately 90% of the cost-adjusted reserve yield on their Hyperliquid supply with the protocol, so the switch gave a venue that creates demand for dollars a recurring claim on the income those dollars generate.

This piece argues that distribution becomes more valuable when it produces persistent, monetizable balances, which depend on three conditions that neither a large audience nor a token balance can meet on its own. Users need a reason to keep money in the product, the platform needs some influence over which stablecoin they use, and an enforceable arrangement has to allocate part of the reserve income to it.

The sections below compare price float against the fee income distributors already earn, examine how Hyperliquid, Phantom, Jupiter, and MegaETH claim reserve income, and then look at what limits that income and how to measure what a platform keeps.

Putting a price on float

What distribution already earns from trading

MetaMask charges a standard swap fee of 0.875%, subject to discounts, and Phantom's published perps schedule lists a 0.05% Phantom component alongside a 0.045% Hyperliquid component, with variations by market, order type, and account tier. In both cases, the interface takes a meaningful share of the transaction while relying on another venue's execution, which already shows that distribution has a price.

DefiLlama records $23.42m of gross protocol revenue across Phantom's tracked products in Q3 2026, up 13.2% from $20.69m in Q2. Its methodology counts wallet fees and Hyperliquid builder-code income, so the series describes a transaction business and excludes both corporate net profit and CASH reserve income. Float would add a second revenue stream to that business without making it obsolete.

Phantom weekly fees and revenue, April to November 2026

  • Fees
  • Revenue
$0$1M$2M$3M20 Apr18 May15 Jun13 Jul10 Aug7 Sep5 Oct2 Nov
Phantom gross protocol revenue, Q2 and Q3 2026.

DefiLlama ↗Data as of 17 Nov 2026

Show the numbers
WeekFeesRevenue
20 Apr 2026$1,384,009$1,384,009
27 Apr 2026$1,374,276$1,374,276
4 May 2026$1,848,542$1,848,542
11 May 2026$1,910,942$1,910,942
18 May 2026$1,568,856$1,568,856
25 May 2026$1,387,476$1,387,476
1 Jun 2026$1,835,544$1,835,544
8 Jun 2026$1,475,658$1,475,658
15 Jun 2026$1,498,687$1,498,687
22 Jun 2026$1,471,384$1,471,384
29 Jun 2026$2,123,568$2,123,568
6 Jul 2026$1,891,595$1,891,595
13 Jul 2026$1,434,631$1,434,631
20 Jul 2026$1,318,760$1,318,760
27 Jul 2026$1,247,175$1,247,175
3 Aug 2026$1,559,887$1,559,887
10 Aug 2026$1,305,572$1,305,572
17 Aug 2026$2,395,907$2,395,907
24 Aug 2026$2,193,524$2,193,524
31 Aug 2026$1,758,840$1,758,840
7 Sep 2026$1,911,907$1,911,907
14 Sep 2026$1,876,125$1,876,125
21 Sep 2026$2,165,305$2,165,305
28 Sep 2026$1,799,627$1,799,627
5 Oct 2026$1,554,435$1,554,435
12 Oct 2026$1,495,653$1,495,653
19 Oct 2026$1,225,635$1,225,635
26 Oct 2026$1,281,192$1,281,192
2 Nov 2026$1,386,265$1,386,265
9 Nov 2026$1,513,810$1,513,810
Embed or cite

Free to reuse under CC BY 4.0: credit Chimera Labs and link to this chart.

Embed code

<iframe src="https://www.houseofchimera.com/embed/charts/phantom-gross-protocol-revenue" title="Phantom weekly fees and revenue, April to November 2026" width="100%" height="420" style="border:0" loading="lazy"></iframe>
<p>Chart: <a href="https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-phantom-gross-protocol-revenue">Phantom weekly fees and revenue, April to November 2026</a> by Chimera Labs, House of Chimera (<a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>)</p>

Citation

Chimera Labs (2026). "Phantom weekly fees and revenue, April to November 2026" [chart], data as of 17 Nov 2026. House of Chimera. https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-phantom-gross-protocol-revenue

Download CSV

The reserves backing a stablecoin keep earning interest whether or not their holders trade, and contracts decide how much of that interest reaches the issuer, the distributor, or the wider ecosystem. The distributor does not need custody to receive a share. In its May 2026 submission to the OCC (Office of the Comptroller of the Currency), Consensys states that MetaMask is non-custodial and that Consensys receives a commercial distribution fee from Bridge, derived from mUSD's reserve earnings, so economic participation lies with the product distributing the stablecoin, while issuance and reserve management are elsewhere.

Float in this piece refers to the reserve-backed balances that generate this income, which the application neither owns nor can draw on to absorb trading losses.

How much a $5bn balance could earn

Annual float income = average eligible balance × cost-adjusted reserve yield × distributor share

Taking the $5bn balance from Coinbase's May announcement as an illustrative constant, a 3% cost-adjusted yield and a 90% share would produce $135m a year, while the same calculation gives $90m at 2% and $180m at 4%. These scenarios are not forecasts of Hyperliquid revenue, since the announcement's figure is an ecosystem-wide reference at scale and not a verified annual average of AQAv2-eligible balances. Actual receipts will depend on daily eligible balances, the applicable reference rate, costs, and activation timing.

Yearly stablecoin float income on a $5bn balance at a 90% share, by reserve yield

$0$50M$100M$150M$200M2% yield3% yield4% yield
Annual float income on a $5bn balance at three yields.

House of Chimera calculationData as of 18 Nov 2026

Show the numbers
YieldAnnual float income
2% yield$90,000,000
3% yield$135,000,000
4% yield$180,000,000
Embed or cite

Free to reuse under CC BY 4.0: credit Chimera Labs and link to this chart.

Embed code

<iframe src="https://www.houseofchimera.com/embed/charts/float-income-by-reserve-yield" title="Yearly stablecoin float income on a $5bn balance at a 90% share, by reserve yield" width="100%" height="420" style="border:0" loading="lazy"></iframe>
<p>Chart: <a href="https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-float-income-by-reserve-yield">Yearly stablecoin float income on a $5bn balance at a 90% share, by reserve yield</a> by Chimera Labs, House of Chimera (<a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>)</p>

Citation

Chimera Labs (2026). "Yearly stablecoin float income on a $5bn balance at a 90% share, by reserve yield" [chart], data as of 18 Nov 2026. House of Chimera. https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-float-income-by-reserve-yield

Download CSV

Circle's rounded reserve return rate of 3.5% for Q3 2026 provides context for the rate assumption, although it is a company-wide figure and is separate from the cost-adjusted rate underlying Hyperliquid's share.

Comparing float with fees comes down to how much trading volume would be needed to earn the same $135m. At Phantom's 5 basis point fee (0.05%), it would need $270 billion in annual transaction volume to generate $135 million in revenue, equivalent to 54 times its $5 billion stablecoin balance. If the fee were just 1 basis point (0.01%), the required annual volume would rise to $1.35 trillion, or 270 times the balance. These figures exclude rebates and operating expenses.

Trading volume needed to match $135m of yearly float income, by fee rate

$0$500B$1T$1.5T5 bps / 0.05%1 bp / 0.01%
Trading volume needed to match $135m of float income

House of Chimera calculationData as of 18 Nov 2026

Show the numbers
FeeAnnual volume needed
5 bps / 0.05%$270,000,000,000
1 bp / 0.01%$1,350,000,000,000
Embed or cite

Free to reuse under CC BY 4.0: credit Chimera Labs and link to this chart.

Embed code

<iframe src="https://www.houseofchimera.com/embed/charts/volume-to-match-float-income" title="Trading volume needed to match $135m of yearly float income, by fee rate" width="100%" height="420" style="border:0" loading="lazy"></iframe>
<p>Chart: <a href="https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-volume-to-match-float-income">Trading volume needed to match $135m of yearly float income, by fee rate</a> by Chimera Labs, House of Chimera (<a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>)</p>

Citation

Chimera Labs (2026). "Trading volume needed to match $135m of yearly float income, by fee rate" [chart], data as of 18 Nov 2026. House of Chimera. https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-volume-to-match-float-income

Download CSV

The comparison explains why balances matter without implying that float always earns more than fees, since a high-turnover trading business can still generate more from execution. A product holding substantial balances between transactions can also earn on the time money spends in it, as well as on the activity.

How platforms claim a share of reserve income

Hyperliquid negotiated a share of USDC

Hyperliquid's arrangement shows how a platform with established liquidity can maintain a widely accepted stablecoin while still participating in its economics. Under AQAv2, reserve income is transferred to the Assistance Fund after specified accrual intervals, and the framework ties the commercial obligation to on-chain transfers and deployer staking commitments.

Seen this way, the wind-down of USDH reads less like a failed branded coin than as a trade in which Hyperliquid maintained an economic claim on its ecosystem's dollars without having to persuade users to adopt another. Its valuable asset throughout was the ability to organise demand around a settlement asset, which the move to USDC left intact.

Phantom built a branded dollar into the wallet

Phantom's approach runs through a branded dollar instead, with CASH's published terms allocating rewards, after issuance fees, to contributors on the supply they originate. Phantom Cash connects that balance to bank funding, transfers, trading and card spending, which gives users several reasons to keep a working balance in the wallet.

Integration of this kind shows that a retention mechanism exists, although proving that retention will last requires balance histories, spending activity and evidence of how users behave once incentives change. Total CASH supply also cannot automatically be treated as Phantom's revenue base, because other contributors can originate supply as well.

Outsourced issuance at Jupiter and MegaETH

Jupiter and MegaETH both use Ethena infrastructure to issue their dollars and then direct the reserve income towards their own ends. JupUSD's documentation states that revenue from its backing assets flows to the Jupiter ecosystem, while MegaETH describes USDm as using reserve yield to fund sequencer operations and support low transaction costs.

The two arrangements show that reserve income can serve quite different purposes, whether supporting an ecosystem, subsidising a service or funding a protocol treasury, and that launching a stablecoin does not by itself give token holders any claim on that income.

The three routes can also overlap, since a branded stablecoin may well rely on an external issuance platform. In every case the economic question is the same, namely which party originates the balance, what share of reserve income it receives and how much it spends to keep that balance in place.

What makes float income last

The retention problem behind open issuance

Bridge's Open Issuance platform offers reserve management, liquidity infrastructure and configurable stablecoin issuance, and Ethena's whitelabel offering similarly separates reserve operations from product design and distribution. Both reduce the operational burden of entering the market but leave customer demand with the distributor, where the outcome turns on whether users benefit enough to keep their funds in the product. A large incentive can attract deposits without creating durable demand, whereas useful payment or collateral integrations may hold onto balances even without the highest advertised yield.

MetaMask's Money Account documentation describes mUSD balances that can earn through third-party DeFi protocols while remaining available to transact. That user-facing APY sits in a separate economic layer from the reserve-derived distribution fee disclosed by Consensys, and treating the two as one would overstate how directly reserve income reaches users.

DefiLlama's mUSD series shows that reserve-related income is already measurable, recording approximately $267,800 in Q2 2026 and $164,600 in Q3, a decline of 38.5%, with each quarter aggregating the monthly observations for April to June and July to September. The adapter tracks yield accrued by the M assets backing mUSD and allocated to its yield recipient, so it covers neither all MetaMask revenue nor Consensys's net distribution income. The series supports the mechanism while showing that access to reserve economics does not on its own ensure that revenue grows.

MetaMask mUSD backing-asset yield tracked by DefiLlama, Q2 vs Q3 2026

$0$100k$200k$300kQ2 2026Q3 2026
mUSD backing-asset yield, Q2 and Q3 2026.

DefiLlama ↗Data as of 18 Nov 2026

Show the numbers
QuarterBacking-asset yield
Q2 2026$267,800
Q3 2026$164,600
Embed or cite

Free to reuse under CC BY 4.0: credit Chimera Labs and link to this chart.

Embed code

<iframe src="https://www.houseofchimera.com/embed/charts/metamask-musd-backing-asset-yield-q2-vs-q3-2026" title="MetaMask mUSD backing-asset yield tracked by DefiLlama, Q2 vs Q3 2026" width="100%" height="420" style="border:0" loading="lazy"></iframe>
<p>Chart: <a href="https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-metamask-musd-backing-asset-yield-q2-vs-q3-2026">MetaMask mUSD backing-asset yield tracked by DefiLlama, Q2 vs Q3 2026</a> by Chimera Labs, House of Chimera (<a href="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0</a>)</p>

Citation

Chimera Labs (2026). "MetaMask mUSD backing-asset yield tracked by DefiLlama, Q2 vs Q3 2026" [chart], data as of 18 Nov 2026. House of Chimera. https://www.houseofchimera.com/labs/research/defi-s-distribution-war-moved-to-the-stablecoin#chart-metamask-musd-backing-asset-yield-q2-vs-q3-2026

Download CSV

What caps float income

Circle's Q3 2026 results show the clearest limit, as average USDC circulation rose 8% year on year while reserve income fell 8% to $658m after the reserve return rate declined by 61 basis points. More dollars produced less reserve revenue, which makes balance growth on its own a poor guide to earnings growth. In the $5bn scenario above, each percentage point of lower cost-adjusted yield removes $45m of annual income at a 90% share, so diversifying away from trading activity introduces a new dependency on the level of interest rates.

Wider access to reserve economics also invites competition, as rival distributors can offer higher rewards or cheaper transactions, and some of the surplus may migrate to users. That favours products that retain balances economically over those offering the largest temporary subsidy.

Hyperliquid's terms rest on bargaining power that a smaller distributor may not have, since it combines market infrastructure, liquidity and control over which quote assets are eligible. A wallet that merely displays USDC may not hold an equivalent position, since its users can switch stablecoins or interfaces readily while the issuer's liquidity and redemption network is difficult to replace.

None of this removes the reserve, redemption and operational risks underneath the stablecoin, as outsourcing issuance reallocates those responsibilities without eliminating them.

Measuring retained income

The most useful test of the thesis is whether a business can convert retained balances into retained cash flow, which means looking at average eligible balances in place of peak supply and at income actually received in place of an advertised reserve yield. Rewards, distribution payments and operating costs then have to come off, because a distributor that passes most of its economics back through rewards may build a useful product on a modest margin, and supply growth alone will not show it.

Token analysis needs one further step: establishing where the income goes. Payments to a company, ecosystem subsidies, treasury receipts, and token purchases each create distinct economic exposures and warrant different treatment in any valuation.

The next meaningful development will be another platform converting its stablecoin distribution into disclosed revenue-sharing terms, with growing eligible balances, recurring receipts and stable retention costs as confirmation. The thesis would weaken if balances required escalating subsidies, if negotiated shares compressed, or if lower rates erased most of the contribution.

Conclusion

Stablecoin distribution is becoming a recurring-revenue business wherever a useful product keeps balances in place and secures a claim on the reserve income behind them. Hyperliquid’s AQAv2 terms show what an established settlement market can negotiate, while CASH at Phantom, mUSD at MetaMask, and the Ethena-issued dollars at Jupiter and MegaETH show similar economics for funding payments, ecosystem development, and cheaper transactions.

That income is less dependable than the scenarios suggest, since Circle’s reserve income fell 8% in Q3 2026 even as average USDC circulation rose by the same margin, and a platform that passes most of the yield back to users can end up with a useful product on a thin margin.

That puts everyday utility and distribution agreements at the centre of product economics for builders, and moves investors’ attention from headline supply towards what a platform actually keeps. Float can become a substantial revenue source, but its value rests on how much the platform retains after competition, incentives and interest rates have taken their share.

Frequently asked questions

How do DeFi platforms make money from stablecoins?

DeFi platforms can earn a share of stablecoin reserve income through distribution agreements based on the eligible balances they originate or retain.

What is stablecoin float income?

Stablecoin float income is the revenue generated from reserve-backed balances and allocated according to agreed revenue-sharing terms.

What determines the profitability of stablecoin distribution?

Profitability depends on average eligible balances, reserve yields, revenue-sharing agreements, and the costs of retaining users and their balances.

Ask a question about this report →

Share

About the author