Bitcoin has no yield.

That isn't a flaw. It's the design. Proof-of-work pays miners for work — not holders for holding. No coupon, no dividend, no staking reward.

From which follows something uncomfortable: every number sold to you as bitcoin yield comes from somewhere else. Not from the bitcoin. From something you hand over in exchange.

This article maps what that something is. And on one of the five sources, we ran the numbers — and got a result we did not expect.

The one table everything else is measured against

This is the front page of Aave, the largest decentralised lending market, on 14 July 2026. Four rows, unedited.

Dollars: 91 percent lent out. Ethereum: 80 percent. Bitcoin: 3.4 percent.

In units: 32,840 wrapped bitcoin sit there. 1,120 are borrowed.

The market isn't dead — it's full. On 1 June 2026 Aave actually had to raise the WBTC supply cap, from 31,800 to 38,200, because it stood at 97.4 percent capacity.

So more and more bitcoin keeps arriving. And nobody borrows it.

The supply rate doesn't depend on how much is supplied. It depends on how much is borrowed. Which is why it sits at essentially zero.

Nobody wants to borrow bitcoin. Everybody wants to borrow against bitcoin.

Those 32,840 WBTC aren't sitting there as a loan. They're collateral — posted to draw dollars. The credit market for bitcoin itself is all but empty, while the one for dollars runs at 91 percent.

Which makes near-zero the market-clearing price of borrowed bitcoin. In a market that functions, grows, and pays decently for other assets.

And that sets the line every other number in this article has to be read against:

Anything above it is payment for a risk — not for lending bitcoin.

The five sources of a bitcoin yield

The question is never "how many percent." It's: which of these five am I handing over, and am I being paid enough for it?

One: option premium — and here we were wrong

The field has filled up. YieldMax (YBIT), Roundhill (YBTC), NEOS (BTCI), Global X (BCCC), Grayscale (BTCC) — and since 16 June 2026, BlackRock, with BITA. Advertised distribution rates run from 15 to over 45 percent.

The mechanism: write call options against your own bitcoin exposure, pay out the premium. What you hand over is the upside. You keep the crash and sell the recovery.

Our assumption was that this must fail catastrophically over a full cycle. Bitcoin's total return is carried by a few extreme upward phases. A strategy that cuts exactly those off shouldn't work.

So we ran the numbers.

Window A: the last twelve months

YieldMax's YBIT is the largest pure bitcoin option ETF with a track record. Inception: 22 April 2024. It writes weekly call spreads on IBIT, BlackRock's bitcoin ETF.

Which makes the fair comparison YBIT against IBIT. Covered call against simply holding.

The covered call won.

Not by accident, but exactly as designed: in a falling market you collect premium while the underlying drops. That's the one state covered calls are supposed to work in. They worked.

Our assumption was wrong for this window.

Window B: the fund's full life

But twelve months is no test of a strategy that sells upside. Because in those twelve months, there wasn't any.

So: the same comparison across YBIT's entire life — from inception to today. This window contains the late-2024 bull run and the collapse that followed.

Twenty percentage points.

Bitcoin essentially round-tripped in this window: IBIT stood at $36.32, rose to roughly $65 — nearly plus 79 percent — and came back to $33.29. Net: minus 8 percent.

YBIT capped that rise, week after week. And collected premium for it.

Two years of premium did not pay for the upside surrendered in a single run.

The cap is permanent. The premium is not.

That's the asymmetry, and it's mechanical. Every capped rise is a slice of NAV that never comes back. The premium you receive for it is a one-off payment.

And here is the actual point

Same strategy. Same data. Two windows. Two opposite answers.

Judge YBIT on twelve months and you see a strategy that beats holding by five points. Judge it on its full life and you see a twenty-point deficit.

Twelve months without a bull run is not a test of a strategy that sells the bull run.

That's the trap anyone falls into who evaluates a product using the metrics the product itself reports.

What sits on the same page

Now the numbers on the issuer's own product page, two screens apart:

• Distribution rate: 46.40 percent (as of 10.07.2026)
• 30-day SEC yield: 2.85 percent
• Twelve-month total return: minus 40.93 percent

The 46 percent and the minus 41 percent have nothing to do with each other. The distribution rate is the most recent payment, annualised. The SEC yield measures net investment income — and explicitly excludes option premium.

So what is being distributed?

That's disclosed too. Issuers must state what portion of a distribution is return of capital — in a notice under Rule 19a-1.

We ran YBIT's full 2026 distribution table: 28 payments, $6.90 per share in total. Of which $3.47 was return of capital.

Just over half of the distributions — 50.3 percent — was a return of the capital invested. Twelve of the 28 payments were over 80 percent.

At Grayscale's BTCC, the product page states it outright: per the 19a-1 notice of 29 June 2026, the portion of the distribution representing return of capital is 100 percent.

(A caveat that belongs here: 19a-1 figures are estimates. The final split only appears on the year-end tax form.)

Return of capital isn't yield. It's your own money coming back — with the share price falling to match.

And you can watch it happen. YBIT's distribution per share averaged 28 cents in January 2026. By July it was 16 cents. A 43 percent decline in six months — because the distribution is a percentage of a shrinking share value.

The number eats itself.

Two footnotes worth knowing

The fund holds no bitcoin. YBIT's holdings are roughly 95 percent US Treasury bills, plus a purchased call option on IBIT. The product page states, verbatim: "The Fund does not invest in or seek exposure to the current 'spot' or cash price of Bitcoin."

And the benchmark in its own factsheet is the S&P 500. Not IBIT. For a fund whose sole purpose is writing options on IBIT.

A benchmark that cannot answer the only question that matters.

Two: credit

The best-known case is the preferred stock of Strategy (formerly MicroStrategy): STRC (variable), STRF, STRD, STRE (10 percent each), STRK (8 percent). Perpetual, with no maturity.

What Strategy's own prospectus material, filed with the SEC, says:

STRC is not collateralised by Strategy's bitcoin holdings and has only a preferred claim on residual assets. Not a bank deposit. Not FDIC insured. No guarantee of return of principal. The dividend rate is adjustable monthly and may be significantly lower.

That isn't criticism from outside. That's the issuer describing its own instrument.

Then the trajectory. STRC went from 11.25 to 11.50 percent in March 2026 — and, per the 8-K filed 29 June, to 12.00 percent effective 1 July 2026.

The purpose of the variable rate is stated on Strategy's own website: it is adjusted "to encourage trading around STRC's $100 stated amount."

Translated: the coupon isn't a return. It's a price stabiliser.

It rises when the market would otherwise push the paper below par. A rising coupon in a falling market is not generosity. It's a price tag on increased risk.

The context in which it rose: Strategy reported a net loss of $12.4 billion for the fourth quarter of 2025. The stock is down roughly 75 percent from its November 2024 high.

And the dividends aren't paid out of operating business — with quarterly revenue around $123 million, that's arithmetically impossible. They depend on raising new capital or selling bitcoin.

So you aren't holding a bond on bitcoin. You're holding a bet that Strategy can keep raising capital.

Three: basis and carry

Buy spot, sell the future, collect the difference. Direction-neutral. Historically 6 to 8 percent a year, with spikes above 20.

Where does the money come from? The Bank for International Settlements studied it — Working Paper 1087, "Crypto carry." The finding: smaller, trend-following investors buy leveraged futures. That demand pushes futures above spot.

The basis trade is payment for holding the other side of retail leverage demand.

That comes from a central bank working paper, not a marketing deck.

No free lunch. The same research found that at leverage of merely ten times — far below what exchanges offer — the futures leg of such a trade would have been liquidated in more than half the months in the sample.

The trade is direction-neutral in outcome. Not on the way there.

Four: token emission

Babylon built the technically cleanest thing in this field. No wrapping, no bridge, no custodian — the bitcoin stays timelocked on the bitcoin chain and secures other networks. Lombard sits on top with LBTC as a liquid wrapper.

And the reward still doesn't arrive in bitcoin. It arrives in BABY, the protocol token.

Reported rates range from under 1 to 8 percent, depending on who you ask. That range is itself the finding: it isn't disagreement about a number, it's disagreement about what's being counted. Nominal token yield, or real value after the token's own price movement.

Three percent in a token that loses half its value is minus forty-eight percent. The nominal figure is true. It's just meaningless.

BABY fell from its early-2026 highs to around $0.0107 in March before recovering above $0.02 — against 8 percent annual inflation and monthly unlocks running to 2029.

And the market has voted. After two years of maximum attention, under half of one percent of all bitcoin sits in this form of bitcoin DeFi. On Ethereum, the figure is around 15 percent.

Five: operations

On 14 July 2026, Pascal Hügli published a conversation with Chris Ritter, Chief Strategy Officer at the Lightning company Zeus, in his newsletter Less Noise More Signal. Title: "This is the Yield Bitcoiners Have Been Waiting For."

The thesis: Lightning gives bitcoin a native yield. Denominated in BTC, no leverage, no token emission, no surrender of custody.

And on one crucial point, the article is right.

Lightning is the only source on this list where the origin of the return is cleanly explainable: users pay fees for routed payments. The node provides liquidity, the node collects the fee. No token emission, no leverage, no rehypothecation, no dependence on a rising price.

Hügli and Ritter also describe something structurally new: self-custody with managed operations. Via the Validating Lightning Signer, the owner keeps the keys while a professional runs the node — the operator can optimise, but cannot take the bitcoin. For banks that's the real breakthrough: they could offer a bitcoin yield without undermining their own custody solution.

The risks are in the article too, openly: execution, channel topology, liquidity management, rebalancing costs, and — on an uncooperative channel closure — capital locked for roughly two weeks. In their words: "The yield may be native. But it is not passive."

And it still isn't a yield on bitcoin.

It's business revenue from operating payment infrastructure — denominated in bitcoin.

Gold has no yield. A goldsmith makes money. The gold in his workshop is inventory, not a yielding asset.

That explains, without strain, every risk the article itself lists. Those are operating risks of a business, not market risks of an asset.

And bitcoin in a payment channel is not in the vault. It sits in a hot multisig contract with timelocks. It hasn't left the bitcoin system — but it has left cold storage.

The yield nobody can verify

And here the article supplies its own sharpest criticism.

It is headlined "This is the Yield Bitcoiners Have Been Waiting For."

It never once states a yield. No percentage. No range. No order of magnitude.

And it says why: Lightning is private by design. Network-wide volume, average routing income, real capital efficiency — none of it is measurable from outside. Hügli writes this down, and quotes Ritter saying plainly that nobody knows exactly how the Lightning yield scales.

You cannot simultaneously not know what the yield is and know that it is the yield everyone has been waiting for.

The article prints both. That speaks well of it — most would have buried the gap. But it doesn't resolve.

And for anyone who works with base rates, that's a remarkable finding: Lightning is the one asset class where a base rate cannot structurally exist. Not because nobody computed it. Because the data never comes into being.

A badly-run Lightning node doesn't issue a press release.

The full table

What remains

None of these sources is illegitimate. Writing options is a recognised business. So is lending. So, emphatically, is running payment infrastructure.

And the covered call worked in a falling market — we checked that against our own assumption, and our assumption lost.

What's wrong is only the word "yield" — because it implies the bitcoin itself is producing something.

It isn't.

You aren't paid for your bitcoin. You're paid for carrying a risk somebody else wants to shed.

Whether that's a good trade depends on the price. And you can only judge the price if you know what you're selling — and which window you're being shown.

And what we're doing next

The comparison in this article covers two years. That is not a cycle.

YBIT has existed since April 2024. BITA since June 2026. Nobody can say today what systematic call writing on bitcoin would have done over ten years — because the products didn't exist then.

The strategy did. And strategies can be computed.

We're building call writing into our engine as a strategy — with tenor, moneyness and coverage as parameters, across the full bitcoin history, against holding, conditioned on regime. The option premiums we reconstruct from realised volatility and the variance risk premium we measured across two and a half years of real options data — median 13.4 percentage points, positive in 93 percent of observations.

Then the answer will be there. For ten years. Not for two.

Frequently asked questions

Does bitcoin have a native yield? No. Bitcoin uses proof-of-work. The protocol pays miners for computation, not holders for holding. There is no coupon, dividend, or staking reward at the protocol level.

Does a covered-call bitcoin ETF beat simply holding? It depends on the window. Over the twelve months to end-June 2026, YieldMax's YBIT was roughly 4.7 percentage points ahead of the plain bitcoin ETF IBIT — in a falling market, the strategy works. Over the fund's full life since April 2024, which includes a bull run, it was roughly 20.7 percentage points behind.

What's the difference between distribution rate and SEC yield? The distribution rate is the most recent payment annualised — it can include option premium and return of capital. SEC yield measures net investment income and explicitly excludes option premium. A large gap between them is a signal to read the 19a-1 notice.

What is a 19a-1 notice? A mandatory US fund disclosure estimating how much of a distribution came from income and how much from return of capital. Return of capital isn't yield — it's your own capital flowing back.

Is bitcoin staking real staking? No. Bitcoin has no proof-of-stake consensus. So-called bitcoin staking locks BTC to secure other networks, and the reward is paid in those networks' tokens — not in bitcoin.

Why does lending bitcoin pay almost nothing? Because almost nobody wants to borrow it. The collateralised lending rate is the market-clearing price of borrowed BTC, and on 14 July 2026 it stood below 0.01 percent on Aave. Of 32,840 wrapped bitcoin supplied there, only 1,120 were borrowed — 3.4 percent utilisation, against 91 percent for USD Coin. Bitcoin is posted as collateral to borrow other assets — it isn't lent out itself.

Is Lightning routing a bitcoin yield? It's business revenue from operating payment infrastructure, denominated in bitcoin. The origin is cleanly explainable, but it isn't passive — and the network is private, so network-wide yield data does not exist.

---

Sources: Product pages and mandatory disclosures from YieldMax (YBIT), Grayscale (BTCC), Roundhill, NEOS, Global X, BlackRock · IBIT price data (TradingView, reference dates 22.04.2024, 30.06.2025, 30.06.2026) · Strategy SEC filings (8-K, FWP, 10-Q) · Aave V3 Core and governance.aave.com · BIS Working Paper 1087 ("Crypto carry") · Pascal Hügli, Less Noise More Signal, 14 July 2026, in conversation with Chris Ritter (Zeus). All figures carry the date they were taken.

Not investment advice. This article describes mechanisms and their sources of risk — not recommendations.

Study the Past — Improve your Future. 🥋