Deep Dive: TRON's Record USDT Lead Over Ethereum - How the Cheapest Rail Won the Largest Flow
1. The Event: A Record Lead in the Largest Stablecoin Float
As of September 1, USDT supply on TRON stood at $91.8B against $73.7B on Ethereum - a lead of roughly $18B, the widest on record. TRON's USDT float first crossed Ethereum's back in early 2025; what is new this month is the margin. The gap is widening rather than narrowing: over the past 30 days TRON's USDT supply grew 2.28% while Ethereum's fell 1.40%.
This is not a daily-volume spike. It is a float migration that has been running for years, and the migration is accelerating rather than pausing.
2. Background: How the Gap Opened
Three things happened in sequence. USDT launched on TRON in 2019 with a cost structure the Ethereum mainnet could not match. Exchanges and OTC desks routed retail-facing withdrawals onto the cheap rail. Payment corridors in cost-sensitive markets standardised on TRC-20, and once a corridor's liquidity, tooling and compliance stack are built for one rail, switching costs rise even if fees converge elsewhere.
| Phase | What changed | Result |
|---|---|---|
| 2019-2021 | USDT issued on TRON; exchanges adopt TRC-20 withdrawals | Retail flow migrates |
| 2022-2024 | Remittance and merchant corridors standardise on TRC-20 | Habitual usage forms |
| 2025-2026 | Ethereum activity moves to L2s; USDT stays on mainnet | Mainnet USDT float contracts |
| Q3 2026 | TRON's USDT float lead over Ethereum hits a record ~$18B | Lead extends |
3. The Data: The Float
The float is where the lead is decided. TRON holds roughly $91.75B of USDT, about half of all USDT in existence, and it took the largest share of last week's net issuance: +$302.7M against Ethereum's +$189.1M.
| Rail | USDT balance | 7-day change | 30-day direction |
|---|---|---|---|
| TRON | $91.747B | +$302.7M | +2.28% |
| Ethereum | $73.697B | +$189.1M | -1.40% |
| BSC | $9.180B | +$10.8K | Flat |
| Solana | $2.904B | +$70.0M | Growing from a small base |
| Arbitrum | $842.1M | +$1.9M | Flat |
| Polygon | $759.2M | -$45.7M | Declining |
4. The Data: The Flow
TRON's Q2 2026 disclosure gives the flow picture: 1.1 billion transactions (up 12% quarter-over-quarter), $2.08T in stablecoin settlement volume, $2.12T in token transfer volume, 16.4 million active addresses and $722M in protocol revenue (up 18% q/q). Total value locked reached $28.15B, up 8.27% from Q1 and 16.9% year-over-year.
| Quarterly metric | Q2 2026 | Change |
|---|---|---|
| Transactions | 1.1B | +12% q/q |
| Stablecoin settlement volume | $2.08T | Recovered q/q |
| Token transfer volume | $2.12T | +5% q/q for top tokens |
| Active addresses | 16.4M | +14% y/y |
| Protocol revenue | $722M | +18% q/q |
| TVL | $28.15B | +8.27% q/q |
| Average block size | 96kB of 2,000kB capacity | +8.5% q/q |
The capacity line is the one to remember. At 96kB against a 2,000kB ceiling, TRON is using under 5% of its available block space. The throughput headroom is roughly twenty-fold before congestion becomes a factor.
5. The Data: The Cost Engine
Cost is the mechanism. A standard USDT transfer consumes roughly 65,000 energy. Burning TRX to pay for it costs about 6.5 TRX; renting the same TRON energy from a marketplace costs about 1.37 TRX. At TRX near $0.3263 that is roughly $2.12 against roughly $0.45 - and an ERC-20 transfer on Ethereum mainnet costs roughly $1.30-2.40 at 8-15 gwei.
| Method / rail | Cost per USDT transfer | USD | Relative to renting TRON energy |
|---|---|---|---|
| TRON, rented energy | ~1.37 TRX | ~$0.45 | 1.0x |
| TRON, JustLend Buy Energy | Up to 64% below burn | ~$0.76 or less | ~1.7x |
| L2 USDC | n/a | ~$0.01-0.02 | ~0.03x |
| Ethereum ERC-20 | ~65,000 gas | ~$1.30-2.40 | ~4.1x |
| Bitcoin | ~140 vB (est.) | ~$0.90-1.70 (est.) | ~2.9x |
| TRON, burned | ~6.5 TRX | ~$2.12 | ~4.7x |
Note what the table does and does not say. L2s are cheaper in absolute terms, but they are cheaper for USDC, not for USDT - and USDT is where the settlement float lives. Within USDT, TRON with rented energy is the cheapest option available.
6. The Data: The Energy Market That Makes It Work
The cost advantage is not a subsidy - it is a market. TRX holders stake for energy and delegate it, renters buy it by the hour or day, and the price clears around 21 sun per energy unit at the cheapest providers (1 sun is one millionth of a TRX). The signal that this market has matured is buried in TRON's own Q2 report: the TRX burn ratio fell 5%, which the network attributed to users staking for energy instead of burning it.
| Energy market signal | Reading | What it means |
|---|---|---|
| TRX burn ratio (Q2) | -5% | Staking and renting displaced burning |
| JustLend energy rental users | 80,000+ | Rental is a mainstream product |
| sTRX liquid staking | 9.73B TRX across 17,000+ addresses | Deep supply of delegatable energy |
| Cheapest observed rate | ~21 sun per energy (~1.37 TRX per transfer) | Liquid, competitive market |
| GasFree cumulative | 7.7M tx, $132.6B, $8.48M saved | Fee abstraction at scale |
GasFree deserves separate mention because it changes who can transact at all. By deducting the fee in the asset being sent - USDT originally, USDD added in August - it removes the requirement to hold TRX. It has now processed 7.70 million transactions and $132.6B in cumulative stablecoin volume.
7. Why Ethereum Is Losing USDT Share
It is not that Ethereum got worse; it is that Ethereum's roadmap moved USDC activity to L2s while USDT stayed concentrated on mainnet. Ethereum now serves roughly 400,000 daily active wallets on mainnet against more than 3.2 million across L2s, and USDC's own balances show the rotation: mainnet USDC rose $369.3M to $47.496B while Base fell $68.3M and Arbitrum fell $63.4M.
| Ethereum tier | Daily users | Stablecoin cost | USDC balance |
|---|---|---|---|
| Mainnet | ~400K wallets | ~$1.30-2.40 (ERC-20) | $47.496B (+$369.3M) |
| L2s (Base, Arbitrum, others) | ~3.2M users | ~$0.01-0.02 | Base $4.210B (-$68.3M); Arbitrum $2.151B (-$63.4M) |
The result is a clean division: Ethereum owns USDC and institutional settlement, TRON owns USDT and high-frequency value movement. Neither is losing on its own terms.
8. What This Does Not Mean
TRON has overtaken Ethereum as a blockchain.
No. It holds more USDT than Ethereum mainnet does - a lead that first opened in early 2025 and is now at a record margin. Ethereum's total ecosystem activity, DeFi TVL and USDC float are far larger.
Cheap fees are the only reason.
Cost started the migration; liquidity, corridor tooling and exchange withdrawal routing keep it there. Cheaper rails existed and did not win this flow.
USDT is leaving Ethereum because something is wrong with it.
USDT's Ethereum float fell 1.40% in 30 days while its TRON float grew 2.28%. That is reallocation toward the cheaper rail, not an exit from the asset.
TRON transfers are free.
They cost about 1.37 TRX in rented energy, roughly $0.45, and about $2.12 if you burn instead of rent.
9. Impact: On Stablecoin Issuers
For Tether, the implication is operational: the majority of USDT float now sits on one rail, and the marginal dollar of issuance lands there too. That concentrates both the benefit (deep liquidity, predictable costs) and the risk (a single rail's resource market or governance becomes systemically relevant to USDT).
- ☐Track the TRON share of weekly net issuance, not just the total.
- ☐Monitor TRON energy market depth as a systemic input to USDT transfer cost.
- ☐Watch whether regulated-issuer requirements (MAS P015-2026, GENIUS Act rules) change where reserves and float sit.
10. Impact: On Payment Firms and Corridors
Payment firms plan capacity against where liquidity already is. A rail that holds $91.8B of USDT at roughly $0.45 per transfer is the default for any corridor where the alternative is a 3-5% foreign-exchange markup plus three-to-five business days of delay. This week's announcements all assumed exactly that: licensed off-ramps for on-chain settlement, yield on idle balances, QR spend from stablecoin balances, and merchant T+0 settlement.
11. Impact: On Cost Planning for Businesses
For a business, the practical consequence is that method choice now dominates chain choice. On a base of 200 USDT transfers a month, renting TRON energy costs about $90 against roughly $424 for burning - a $4,000 annual difference on the same chain.
| Scenario | Monthly (200 transfers) | Annual |
|---|---|---|
| Rent TRON energy (~$0.45) | ~$90 | ~$1,080 |
| JustLend Buy Energy (~$0.76) | ~$152 | ~$1,824 |
| Ethereum mainnet ERC-20 (~$1.85) | ~$370 | ~$4,440 |
| Burn TRX (~$2.12) | ~$424 | ~$5,088 |
12. What Could Reverse It
Float leads built on cost structure are durable; promotional ones are fragile. This one is mostly structural, with three credible reversal candidates.
| Reversal candidate | Mechanism | Current risk |
|---|---|---|
| TRX price spike | Energy is priced in TRX; a sustained rally raises USD transfer cost | Medium - TRX ~-2.5% w/w, ~25% below its ATH |
| Energy supply squeeze | Staking ratio falls, rental prices rise | Low - 9.73B TRX staked via sTRX alone |
| USDT grows on L2s | Issuer pushes float to cheap L2 rails | Low-Medium - USDT float on Base is minimal today |
| Regulatory action on TRC-20 flows | Corridor restrictions in a major market | Low - none observed this week |
13. Outlook: September and Beyond
If the current composition holds, TRON's USDT float should keep taking the majority of net issuance and the float gap should widen through September. The variables to watch are TRX price - every 10% move shifts the dollar cost of a rented-energy transfer by the same amount - and whether regulated-issuer rules push float toward different rails.
14. Conclusion
TRON did not win the USDT float with a marketing campaign. It won it by being roughly four times cheaper per transfer for years, by having twenty-fold block-space headroom, and by building a real market for the TRON energy that makes that cost possible. Ethereum chose a different path - premium settlement on mainnet, mass-market USDC on L2s - and is not weaker for it. The split is now clear: USDT settles on TRON, USDC settles on Ethereum, and the cost of each follows the rail it chose.
- TRON's USDT float lead over Ethereum hit a record of roughly $18B: $91.8B vs $73.7B, with its float up 2.28% over 30 days while Ethereum's fell 1.40%.
- The mechanism is cost: ~1.37 TRX (~$0.45) for rented TRON energy versus ~$1.30-2.40 for an ERC-20 transfer on mainnet.
- The TRON energy market is real and liquid - 80,000+ rental users, 9.73B TRX in sTRX, and a 5% fall in the burn ratio as staking displaced burning.
- GasFree removed the last friction: 7.7M transactions and $132.6B settled without the sender holding TRX.
- Reversal candidates are watchable but not imminent; the most live variable is TRX price, which transmits directly into dollar transfer cost.
Frequently Asked Questions
Does this mean TRON is bigger than Ethereum?
No. It means TRON holds more USDT than Ethereum does - a lead that first opened in early 2025 and has now stretched to a record margin of roughly $18B. Ethereum's ecosystem activity, DeFi TVL and USDC float remain far larger, and this report compares one metric - USDT float - not overall network size.
How much does a TRON USDT transfer actually cost?
About 1.37 TRX if you rent the ~65,000 energy required, or roughly $0.45 at TRX $0.3263. Burning instead costs about 6.5 TRX (~$2.12), and a first transfer to a brand-new address needs about 131,000 energy.
What is TRON energy and why does it matter?
It is the resource consumed by smart-contract operations such as USDT transfers. You can obtain it by staking TRX or rent it from a marketplace. Because it is priced in TRX, the dollar cost of a transfer moves with the TRX price even when the energy requirement is unchanged.
Aren't L2s cheaper than TRON?
Yes for USDC - about $0.01-0.02 versus ~$0.45. But USDT settlement concentrates on TRON, and within USDT, TRON with rented energy is the cheapest option available.
Is this report financial advice?
No. It is independent research on public data.
Sources & Methodology
This report is compiled from public on-chain data, official announcements and a curated source whitelist. Figures are cross-checked where possible; estimated or reference values are labelled as such. Nothing in this report is financial advice.
- Official project documentation, blog posts and GitHub repositories
- On-chain data from public explorers and analytics dashboards
- Primary announcements from the parties involved
- Cross-checked industry media coverage
Last reviewed: 2026-09-07.