Bitcoin's Carry Trade Is Back After 67 Days in the Red, and It Might Beat T-Bills
A bank desk expects renewed tap into demand to revive the Bitcoin carry trade in 2026, after 67 straight days of negative funding gutted the spread. The boring trade that doesn't need price to rise could out-yield Treasury bills by 2027. Here's the case for it, the case against it, and where I land.
The best trade in crypto doesn't care whether Bitcoin goes up. It spent most of 2026 being a lousy trade anyway. Sixty-seven consecutive days of negative funding, and everyone running spot-and-short got paid in tuition.
Now a bank desk is calling the turn. The expectation is simple. use demand comes back, funding flips positive, and the carry trade starts paying again, enough to beat Treasury bills by 2027. That's the whole thesis. Not price. Spread.
Here's my claim up front. The 67-day stretch wasn't a warning about Bitcoin. It was the cleanup that makes the next carry window worth taking seriously.
The Trade Nobody Brags About
If you've never run it, the mechanics are almost embarrassingly dull. You buy spot Bitcoin. You short the perpetual future against it. Delta goes to zero, so price stops mattering, and what you collect is the funding rate, paid every few hours by whichever side of the perp is crowded.
When longs are greedy, longs pay shorts. You're short. You get paid.
That's it. No charts, no liquidation hunts, no conviction plays. Just a yield that shows up as long as the crowd stays levered long.
In a hot market that funding can annualize in the double digits. In a flat market it compresses toward zero. In 2026 it went negative, which means the shorts were paying the longs, which means the carry trade was paying out instead of collecting. For 67 days. Ask anyone who ran this in a crowded quarter why they stayed. They'll laugh, then they'll tell you about the borrow costs.
But the trade didn't die. It just got cheap to enter. And that's the part the bank desk is pointing at.
Sixty-Seven Days Leaves a Mark
Negative funding for that long tells you something specific about positioning. It says the perp market was leaning short, hard, for two months straight. Traders were paying to be bearish. That's not normal, and it's not cheap.
So what happens when that pressure releases? Two things, and they compound.
First, the shorts who've been collecting funding start closing. That's a cost they were happy to carry while it lasted, and once it flips they're just paying rent on a position they don't need. Second, the spot side gets cheaper to hold because the crowded short caps upside reflexively. When both of those unwind, funding doesn't just go to zero. It tends to overshoot.
That overshoot is the trade. Not the steady state. The snap back.
And there's a structural piece here that the 2021 cycle didn't have. Spot Bitcoin sits in regulated wrappers now. Pension money, RIA money, model portfolios. That capital doesn't want perp exposure. It wants the basis, or it wants nothing. So the short leg of the carry trade has a deeper, stickier supply of counterparties than it used to. That's a real change, and it's why a bank desk is comfortable modeling this out to 2027.
Which brings up the question I keep getting. If this trade is so good, who's on the other side?
Answer. Somebody who wants use and can't get it anywhere else. That's not a bug. That's the product.
Where This Breaks
Now steelman the bear case, because it's a good one.
The obvious risk isn't Bitcoin falling. Delta-neutral means you don't care. The risk is the plumbing. Counterparty failure, exchange insolvency, a custodian that turns out to be a marketing department with a vault. Anyone who lived through 2022 knows exactly how that movie ends, and it doesn't end with your basis trade paying out.
There's a second risk that gets less airtime. Crowding. If the carry trade beats T-bills by a wide enough margin, capital floods in, and the trade's own popularity compresses the yield until it doesn't beat anything. That's what happened to every basis trade in every market ever. It works until the spread is arbitraged to nothing, then it becomes a parking spot with extra steps.
And a third, quieter risk. Treasury yields. The entire pitch is relative. If bills are paying 4 percent and carry is paying 6, you take carry. If bills go to 5 and carry sags to 4.5 because funding never really recovers, you've taken on exchange risk for negative excess return. That's a bad trade dressed up as a smart one.
So the bull case leans on a funding recovery that hasn't happened yet, in a year where the perp market spent two months telling us the opposite.
Fair. That's the honest counterpoint.
The Verdict
I'll take the other side of the bears. Here's why.
Sixty-seven days of negative funding isn't evidence that carry is dead. It's evidence that use got washed out, and washed-out use is the precondition for a funding snap, not the end of one. The desk isn't predicting a bull market. It's predicting that the crowd gets bored of paying to be short.
That's a much easier call to make.
My bigger take, and the one that'll annoy people. The fact that a delta-neutral trade can credibly beat Treasury bills tells you more about bills than about Bitcoin. A 4 percent risk-free rate is a low bar. Clearing it shouldn't require a bank note and a two-year model. It should be table stakes. The interesting thing isn't that carry can win. It's that so few crypto strategies can.
Which is exactly why this one keeps coming back. Behind every protocol is a person who bet their twenties on it, sure. But behind every fund that survived 2022 is somebody who ran the boring trade while everyone else was running the fun one.
Who wins if funding recovers? The desks with spot inventory and cheap borrow. The custodians holding the long leg. The exchanges collecting fees on both sides, forever, in any direction.
Who loses? The trader who hears "carry trade" in July 2027, after the spread has already compressed, and enters at the top of the crowding curve. That person always shows up. They showed up in 2021, they'll show up again.
The window, if the desk is right, opens somewhere between now and 2027. It won't stay open long. These things never do.
And the honest answer to whether I'd put money behind it? I'd put it behind the plumbing first. The spread is the easy part. The custodian is the trade.
Explore More
Key Terms Explained
The first cryptocurrency, created in 2009 by the pseudonymous Satoshi Nakamoto.
A sustained period of rising prices and positive market sentiment.
Borrowing in a low-interest-rate asset to invest in a higher-yielding one, profiting from the difference.
A DeFi lending protocol on Ethereum where you can supply assets to earn interest or borrow against collateral.