On Binance the same coin often has two perpetuals: one settled in USDT and one settled in USDC. Both track the same index, but they trade in separate order books, and each settles funding on its own premium. So the readings differ.
The short answer first. Taking the 34 matched pairs in our database (every USDC perpetual and its USDT twin) and summing each contract’s settlements from 1 January to 27 September 2026:
- In 33 pairs the USDC contract collected more. NEO is the only exception.
- The median gap is +0.90 percentage points.
- For BTC the gap is just 0.12 points and for ETH 0.15: 12 and 15 USDT on 10,000 USDT of notional, earned over nine months.
The big gaps are all in altcoins, which is exactly where the other costs are highest. Here are the numbers.
What 34 pairs paid this year
| Coin | USDC perp, cumulative | USDT perp, cumulative | Gap (points) |
|---|---|---|---|
| BTC | 2.282% | 2.161% | +0.12 |
| ETH | 1.534% | 1.387% | +0.15 |
| LINK | 4.019% | 3.466% | +0.55 |
| SOL | −0.187% | −0.865% | +0.68 |
| DOGE | 3.155% | 2.174% | +0.98 |
| XRP | 1.483% | −0.233% | +1.72 |
| SUI | 3.923% | 2.061% | +1.86 |
| NEO | −3.530% | −1.086% | −2.44 |
Only eight pairs are shown: BTC and ETH, plus the ones that make the point most clearly. Across all 34, the median gap is +0.90 points.
Two things stand out:
- For XRP and WLFI, the sign itself differs. XRP’s USDC perpetual is at +1.48% for the year while the USDT one is at −0.23%; WLFI is +0.44% against −0.32%. Same coin: a short on one contract has been collecting all year, a short on the other has been paying.
- Where both are negative, the USDC contract is less negative. BIO is −9.53% on USDC and −13.58% on USDT; KAITO is −30.91% against −33.52%. It looks as if the USDC readings are shifted up a little across the board, rather than a few coins happening to run high.
37,144 settlements at the same instant
Cumulative totals smooth over the detail, so we lined up the two contracts at the same settlement time, period by period. The 34 pairs give 37,144 matched settlements:
| At the same settlement | Periods | Share |
|---|---|---|
| USDC contract higher | 16,878 | 45.4% |
| Identical | 9,861 | 26.5% |
| USDT contract higher | 10,405 | 28.0% |
The USDC contract was higher 1.6 times as often as the USDT one. The direction is clear, but it is far from every period. One more number matters more: in 8,085 of those periods (21.8%) one contract was positive and the other negative. One was paying shorts while the other was charging them. They are not two copies of the same income.
BTC and ETH: a gap small enough to ignore
BTC’s two contracts each settled 808 times this year, every eight hours, and the most common reading on both was 0.010000% (67 times on USDC, 56 on USDT). At matched times, USDC was higher 409 times, USDT 374 times, and 25 were identical; 149 had opposite signs. ETH was 409 against 378, with 21 identical and 154 of opposite sign.
Gaps of 0.12 and 0.15 points come to 12 and 15 USDT on 10,000 USDT of notional. Chasing that means closing one perpetual, opening another and moving the spot leg from USDT to USDC pricing, each step with fees and slippage. On the majors it is not worth doing. The costs that never show up on screen are listed in the six hidden costs of funding rate arbitrage.
Why the USDC contract runs slightly higher
Each contract settles funding on its own premium: when the perpetual trades above the index, longs pay; below it, shorts pay. The formula and the three most common mistakes are in how funding rates are calculated.
Our data shows what happens, not why. One plausible explanation: the force that pushes a premium back is mostly arbitrage capital (buy spot, short the perpetual), and most of that capital posts USDT as margin. The USDT perpetual therefore gets more shorts and a flatter premium; the USDC book gets less of that pressure, keeps a little more premium and pays a little more funding. That fits the 33-to-1 split, but we do not have positioning data for either book to prove it, so treat it as a hypothesis.
Four things to price before moving to USDC contracts
- Is there a USDC spot pair? Both legs need the same pricing currency to be balanced. BTC and ETH have USDC spot books on Binance; many smaller coins do not.
- USDC margin adds a stablecoin layer. USDT and USDC can each drift briefly from one dollar, and the gap between them lands directly on account equity. This year’s actual deviations are in does a delta-neutral position stay neutral when a stablecoin depegs.
- Order book depth. USDC perpetuals generally trade less than their USDT twins, so the same size costs more slippage. We have no data on this in our database, so this is a warning without a number.
- “Short the USDC perp, long the USDT perp” is not a neutral trade. Both legs are perpetuals, both settle, and 21.8% of periods had opposite signs. It is a bet on the premium spread between two order books, not a collection of funding. USDT- versus coin-margined arbitrage works through a similar question about what “neutral” means once the collateral changes.
Bottom line
- BTC and ETH: the two contracts pay almost the same, 0.12 to 0.15 points apart this year. Pick the one that matches your margin currency, not the one with the higher rate.
- Altcoins: USDC perpetuals really do run slightly higher, but the extra often does not cover their liquidity cost and stablecoin risk.
- On any funding leaderboard, the same coin can appear twice with different readings. Check which contract you are looking at before comparing.
Our yearly performance figures use BTCUSDT and ETHUSDT only; the numbers, worst year included, are on the performance page, and how the two legs are run is on How it works.
Data and method: per-period funding rates come from Binance’s public USD-M endpoint fapi/v1/fundingRate, via our collection database, from 2026-01-01 00:00 UTC to the last period collected at 2026-09-27 01:00 UTC. The 34 pairs are every USDC perpetual in the database together with its USDT counterpart; USDC perpetuals were added from August 2025 and all 34 pairs have data for the whole of 2026. Cumulative figures are sums of periods, not compounded. “Same settlement” pairs readings whose timestamps are within 5 seconds of each other. For ADA, HBAR and ZEC the two contracts settle on different intervals (USDC every 4 hours, USDT every 8 hours); cumulative totals are summed over each contract’s actual settlements and remain comparable, and period-by-period comparisons use only times when both settled. Tickers are shown to document the data, not as recommendations. Past data does not indicate future returns.