ETH Supply in 2026: Buterin Rethinks the Burn, Staking Peaks

Ethereum spent four years telling investors a simple story: every transaction burns a little ether, the burn offsets new issuance, and over time the supply shrinks. On Sunday, November 8, 2026, the person who did more than anyone to popularize that story told holders to stop leaning on it. In a morning post on X, Vitalik Buterin argued that ETH makes far more sense as a store of value than as a fee-burning token, and that investors should not feel sad if the burn stays small for good.

Two days later, on-chain data filled in the other half of the picture. A record share of all ETH, more than a third of the supply, is now locked in staking, an all-time high reached while the price was sliding toward $2,000. Read together, the two events redraw the supply question. The automatic scarcity engine that justified the ultrasound-money meme is effectively retired. What remains is a demand-side lock, staking plus corporate treasuries plus exchange-traded funds, that keeps pulling ether off the market even as the economics of holding it get harder to defend. This piece lays out where ETH supply stands in November 2026, why the burn stopped mattering, and whether the lock can hold into 2027.

Buterin Retires the Burn as the Price Case

Buterin was careful to separate the mechanism from the marketing. He wrote that EIP-1559, the 2021 change that burns part of every transaction fee, is “fine and good” and that the network should keep it. His argument was about how to value the asset, not about the code. The case for pricing ETH as a “fee-burning token,” he said, was always weaker than it looked during the last cycle, and treating “ETH as store of value” makes “much more sense to focus on.” The line that traveled fastest was his conclusion that the market “should not feel sad if fee burn is < 10% of breakeven indefinitely; the value of ETH comes from elsewhere” (The Block).

The numbers behind the shift are stark. Buterin noted that Ethereum burned roughly 1.48 million ETH in 2022, worth about $3.6 billion at the time, and only around 19,200 ETH, close to $43 million, in all of 2026 so far. Against a market value near $270 billion that weekend, a burn that small is a rounding error. His point was less that the burn is broken, a fact the market has lived with for two years, and more that no honest valuation of ETH can lean on it. Discount the stream of burned fees the way you would any cash flow and the present value sits far below where ETH trades, so whatever justifies the price has to come from somewhere other than the burn.

Where, then? Buterin put the weight on Ethereum’s harder-to-copy properties: decentralization, censorship resistance, privacy, and security, adding that in “the era of AI disruption, stability will be highly valued.” It is a Bitcoin-style argument dressed in Ethereum’s clothes, and it lands at an awkward moment. A store-of-value case depends on holders locking ether away and keeping it there. The data say they are doing exactly that, at a record pace, which is where the supply story gets interesting.

The Supply Snapshot

Before the arguments, the levels. Here is where ETH supply sits in the middle of November 2026, drawn from on-chain trackers and fund data.

Two things jump out. First, the protocol is growing the supply, not shrinking it: net issuance has been modestly positive since the Dencun upgrade of March 2024, the opposite of the deflation the ultrasound-money crowd expected. Second, nearly every bucket that could absorb that new supply is doing so, with staking at a record and only a thin slice of ether actually free to trade. The rest of this piece works through that tension, starting with why the chain mints more than it burns.

Why the Chain Still Mints More Than It Burns

Ethereum has two sources of new ether and one sink. Validators who secure the network earn freshly minted ETH, currently about 2,970 coins a day, a little over a million a year, scaled to the amount staked (ultrasound.money). Against that, EIP-1559 burns the base fee on every transaction. When the network is busy and fees are high, the burn can exceed issuance and the supply falls; when fees are low, issuance wins and the supply grows. For most of 2022 and 2023 the burn won often enough to make ETH mildly deflationary. Since Dencun, it has not come close.

The culprit is Ethereum’s own scaling success. The Dencun upgrade in March 2024 introduced blob space, a cheap data lane for layer-2 rollups, and activity promptly migrated there. Rollups now settle the bulk of transactions for a fraction of a cent, which is wonderful for users and terrible for the burn, because the fees that once got destroyed on the main chain are now tiny blob fees that barely register. On a typical day in November the network minted close to 3,000 new ETH and burned a few dozen. For the burn to offset issuance again, the base fee would have to rise roughly three-hundredfold, back toward the double-digit gwei levels of the last bull market, and stay there. More than 4.6 million ETH has been burned since EIP-1559 went live in August 2021, but the pace today is a trickle, and nothing on the roadmap is built to reverse it.

If Not the Burn, Then What?

Buterin’s post reframes a question analysts have circled all year: if ETH is not a deflationary, bond-like instrument, what is it? The most-cited answer from the sell side treats ether less as money and more as a claim on the network’s activity.

Karim AbdelMawla, a senior researcher at the issuer 21shares, has argued that ETH “increasingly behaves as a leveraged claim on ecosystem activity rather than a pure monetary asset,” and that lower fees shrink the burn unless on-chain activity grows enough to make up the difference (21shares). In that frame the burn is not a scarcity engine at all; it is a dividend that only shows up when blockspace is in heavy demand. The valuation exercise then looks like any discount-rate problem, the same present-value discipline that decides whether, say, a verifiable-AI network earns its hurdle rate: estimate the cash flows the asset throws off, pick a discount rate, and check whether the number supports the price.

Run that exercise on the burn alone and ETH looks wildly overvalued, which is exactly Buterin’s point and exactly why he wants the conversation moved to store-of-value ground, where the inputs are softer and the comparison is Bitcoin rather than a bond. The catch is that a store-of-value asset has to prove its scarcity in the market, through holders who refuse to sell, rather than in the protocol, through a mechanism that destroys supply. That shifts the whole burden of the bull case onto the demand side, and onto one question in particular: who is locking ether away, and will they keep doing it?

Scarcity Moved to the Demand Side

If the protocol will not manufacture scarcity, the market has to. Three pools of demand have been doing that work in 2026: validators who stake, companies that hold ETH as a treasury asset, and the spot ETFs that custody ether for investors. Each pulls coins out of the tradable float, and in the staking and treasury cases tends to keep them out for long stretches.

One caution before adding these up: the buckets overlap, so they are lenses, not line items. BitMine, the largest corporate holder, stakes most of its ether, so the same coins appear in both the treasury total and the staking total. A fund that passes staking rewards through to investors shows up in both the ETF tally and the staking tally. Summing them would overstate how much supply is truly locked. The cleaner way to see the squeeze is from the other side: how little ether is left where it can be sold.

That number keeps falling. Ether held on exchanges has slid to multi-year lows, and the coins that still change hands do so through the automated market makers and order books that set the price on increasingly thin volume. A thin float cuts both ways. It amplified the summer rally when buyers arrived, and it amplified the autumn slide when they left. Scarcity, in short, is now a market condition rather than a protocol guarantee, and market conditions reverse.

Staking Just Hit a Record, and That Is the Real Story

The biggest of the three locks is staking, and it just set a record. On November 10 the share of ETH committed to validators reached 36.3 percent of supply, an all-time high, roughly 44 million coins (TokenPost, citing CryptoQuant). What makes the figure notable is the backdrop: it climbed while ETH was falling toward $2,000. Holders were locking up more supply, not less, into a declining price. Bulls read that as conviction; bears read it as a crowded trade waiting to unwind.

There is a wrinkle. The rate of new staking is slowing. Validator inflows fell by about 40 percent, and roughly 900,000 ETH sat in the entry queue in mid-November, a fraction of the multi-million-ETH backlog that formed in late spring when yield-bearing ETFs and treasury buyers were piling in. The stock of staked ether is at a record; the flow into it is cooling. That distinction matters, because the climb toward ever-higher staking has been the most reliable supply story of the last two years, and a stall in the inflow is the first thing that would change its character.

Who does the staking shapes the risk, too. Lido, the largest liquid-staking protocol, still accounts for a bit over a fifth of all staked ETH and the bulk of the liquid-staking market, even as its share has drifted down while exchanges, ETFs, and treasuries take on more of the work; the demand problem facing Lido, Rocket Pool, and Frax is a story in itself. The trajectory from here is the open question. Projections that assume the trend simply continues put well over half the supply behind the validator lock within two years. That would be a remarkable scarcity signal and, at the same time, a concentration and liquidity risk, because staked ether cannot be sold at a moment’s notice.

The Yield Squeeze: Paid Less Than Treasuries

Here is the uncomfortable part of the record. The reward for locking ether up is at its weakest in years, and it now loses a straight comparison to risk-free cash. The base consensus reward has drifted to about 2.6 percent, a three-year low, as more validators crowd in and split the same issuance; adding MEV through MEV-Boost lifts a well-run validator to roughly 3.3 to 3.8 percent (the mechanics of that validator paycheck are worth their own read). A 10-year US Treasury, meanwhile, pays about 4.8 percent with no lockup and no smart-contract risk (24/7 Wall St).

On a $10,000 stake held for a year with the ETH price flat, staking returns about $300 while the Treasury returns roughly $484, a gap near $184 that compounds the longer it runs. So why does a record share of supply stay staked? Because the yield is beside the point for most of the holders doing it. The reward is paid in ether, which means the real return is dominated by the price of ETH, not the staking rate: a 10 percent move in the token dwarfs years of the yield gap. For a treasury company or a staking ETF, staking is a way to earn something on an asset they were going to hold anyway, not a rate they are shopping against T-bills. The yield is a boost, not the thesis.

That is fine as long as the price cooperates. The danger is the reflexive version: a holder base that stakes for price upside rather than yield is a holder base that could unstake quickly if the price thesis cracks, and the exit queue, not the yield, would then set the pace of any unlock. For now the exit queue is near empty and the entry queue is still positive, so the lock holds. It is simply holding for reasons that have little to do with the number on the staking dashboard.

BitMine Is About to Stop Buying

The second lock, corporate treasuries, is dominated by one name, and that name is about to go quiet. BitMine Immersion Technologies, chaired by Fundstrat’s Tom Lee, held about 6.07 million ETH as of November 8, roughly 4.97 percent of all supply, after adding another 9,587 coins the week before (The Block). That leaves it around 49,500 ETH short of a self-imposed ceiling of 5 percent of supply. More than 86 percent of the hoard, some 5.2 million ETH, is staked, which the company projects will throw off about $359 million a year in rewards once fully deployed.

Lee has been explicit that the buying stops at the ceiling. Speaking at the Token2049 conference in October, he said BitMine “only need to get another 100,000 ETH to get to 5%. Now we’re going to stop,” adding that “we thought this would take five years. It took us a little over a year” (CoinDesk). He stays loudly bullish, pitching ETH as the base layer for a tokenization and AI supercycle, but the distinction matters for supply: a bullish holder who has stopped buying is no longer pulling new coins out of the float.

For most of the past eighteen months, BitMine was a steady, price-insensitive bid, buying through rallies and dips alike. Take that bid away and the market loses its most predictable sponge for new issuance right as ETF demand wobbles. The ether BitMine already holds stays locked, so the stock of scarce supply does not shrink; what changes is the flow. The single largest force pulling new coins out of circulation is, by its own plan, about to switch off.

The ETF Bid Went Into Reverse

The third lock, the spot ETFs, has stopped adding and started subtracting. US spot Ether funds bled about $144.8 million on November 11, a fifth straight session of outflows, trimming their combined net assets to roughly $11.65 billion, about 4.6 percent of ETH’s market value (COINOTAG). BlackRock’s ETHA, the biggest of the group, took the largest share of the November 11 redemptions even as it stays the category leader by assets.

The reversal is the telling part. For most of 2026 these funds were a one-way sponge, and the regulatory table was set for exactly that: US regulators cleared staking rewards as not securities earlier in the year, a joint read from the SEC and the CFTC that let American funds and treasuries stake at all and turned the ETF wrapper into a supply sink with a yield attached. When the flows run in reverse, the wrapper works the other way, releasing custodied ether back toward the market. The ETFs did not build a slow structural bid so much as a fast, sentiment-driven one, and sentiment in November has been poor.

No Help Coming From the Protocol

Could Ethereum just change the rules to manufacture scarcity? It considered exactly that, then backed away. EIP-8363, a proposal to burn a rising share of validator rewards until net issuance hit zero near 60 million staked ETH, spent two months as the most-debated idea in Ethereum governance before its authors pulled it from the Hegota upgrade on October 1 (The Block). Builders had lined up against it, arguing that cutting the staking reward would punish the network for its own growth, and core developers agreed that a fork-scoping process was the wrong venue for a change to monetary policy.

The proposal is not dead; it has been routed into a dedicated issuance-policy process that runs through 2027, with an issuance forum at Devcon this month and a final session penciled in for EthCC in Cannes next April. But routed into a multi-year process is, for supply purposes, much the same as shelved. There is no mechanism coming in the near term to cut issuance or revive the burn. Ethereum’s monetary settings are frozen in their mildly inflationary configuration, which throws the entire scarcity question back onto the demand-side lock this piece has been tracking.

Glamsterdam Is a Scaling Fork, Not a Scarcity Fork

The next upgrade will not help either, and may quietly hurt. Glamsterdam, Ethereum’s major 2026 hard fork, activated on the Sepolia test network on October 6 but then slipped: a planned testnet activation was postponed in late October after auditors flagged risks in one of its gas-repricing changes, and the mainnet date, once floated for early December, is now uncertain, with December 1, 8, and 9 all circulating as candidate slots (Cryptopolitan).

When it does land, Glamsterdam is a scaling release, not a monetary one. Its headline items, enshrined proposer-builder separation and block-level access lists, make block production fairer and validation faster, and a companion change aims to lift the gas limit toward 200 million from 60 million. More capacity is good for users and for the network’s long-term relevance. It is not good for the burn. EIP-1559 targets blocks that are half full, so bigger blocks generally mean lower fees per unit of demand, which means less ether destroyed, not more. The scaling roadmap that makes Ethereum useful is the same roadmap that keeps the burn suppressed. Glamsterdam adds capacity; it does not add scarcity.

ETH Versus Bitcoin: Scarcity by Behavior, Scarcity by Protocol

Buterin’s pivot invites the Bitcoin comparison, and the comparison is unflattering in one specific way. Bitcoin’s scarcity is written into the protocol: the supply schedule is fixed, the halving is automatic, and no committee votes on it. Holders do not have to do anything for new issuance to fall; it falls on a clock. Ethereum’s scarcity, after the retreat of the burn and the shelving of EIP-8363, is the opposite kind. It is produced by behavior, by validators who stake, companies that hoard, and funds that custody, and behavior can change.

That is the quiet significance of the November 8 post. By moving ETH’s story from fee-burn scarcity to store-of-value scarcity, Buterin is asking the market to value Ethereum the way it values Bitcoin, on credibility and conviction rather than on a destruction mechanism. It is an honest reframing. It is also a harder sell, because it asks holders to supply the scarcity themselves, and to keep supplying it through a stretch where the yield is thin, the biggest buyer is capping out, and the funds are redeeming. Scarcity by protocol does not care about any of that. Scarcity by behavior cares about all of it.

The Rate Wall Next Door

None of this plays out in a vacuum. The competition for a holder’s capital is a US Treasury yielding close to 5 percent, and the direction of that yield just got worse for ETH. October inflation, reported on November 10, came in hot, with headline consumer prices up 0.4 percent on the month and core up 0.3 percent, both above forecasts, and futures promptly priced about an 85 percent chance that the Federal Reserve raises rates again at its December 9 meeting, with a minority betting on a larger half-point move (TokenPost). A higher risk-free rate widens the gap that staking already loses, and it is the macro question that narrows the whole crypto calendar to December.

For the supply thesis, the rate wall is the stress test. A holder base that stakes for price upside can shrug off a thin yield as long as it believes in the price; it is much harder to shrug off both a thin yield and a falling price while cash pays nearly 5 percent next door. If December brings another hike and the ETH price keeps sliding, the reflexive risk in the staking lock goes up, because holders who staked for upside are exactly the ones most likely to head for the exit queue if the price keeps falling. The US regulatory frame is settled enough that staking ETFs and treasuries can keep operating under the SEC’s current posture; the open variable is not the rules, it is the rate and the tape.

What to Watch Into 2027

Pull it together and the 2026 supply story is not about whether ETH is scarce; by most measures it is, with a record share staked and a thin tradable float. The story is about what kind of scarcity it is, and whether a demand-side lock can stand in for a protocol one. Here are three ways the next several months could run.

Five dials tell you which way it is breaking. First, the staking ratio: another leg up confirms the lock, while a roll over from the record is the first crack. Second, the validator exit queue: empty means holders are staying, a sudden build means they are leaving. Third, the treasuries: whether companies like BitMine that have stopped buying start selling. Fourth, ETF flows: whether the November outflow streak is a blip or a trend. Fifth, the Fed: each hike raises the bar that staking has to clear. Watch those five and you are watching the whole supply question, because in 2026 Ethereum’s scarcity stopped being something the protocol guarantees and became something its holders decide, day by day.

Frequently Asked Questions

Is Ethereum deflationary in 2026?

No. As of November 2026 Ethereum issues roughly 2,900 to 3,000 new ETH a day to stakers and burns only a few dozen, so the supply is growing at a modest pace of a few tenths of a percent a year. The network was briefly deflationary between 2022 and early 2024, but the Dencun upgrade pushed activity onto layer-2 networks and collapsed the fee burn.

What did Vitalik Buterin say about ETH on November 8, 2026?

In a post on X, Buterin argued that ether makes more sense as a store of value than as a fee-burning token, and that holders should not feel sad if the burn stays far below what would be needed to offset issuance. He said the value of ETH comes from decentralization, censorship resistance, privacy, and security, and that the EIP-1559 burn is fine and should stay.

How much ETH is staked right now?

About 44 million ETH, a record 36.3 percent of total supply, is staked as of mid-November 2026. The share hit an all-time high even as the price fell toward $2,000, which many analysts read as a sign of long-term holder conviction rather than short-term trading demand.

Why do people still stake ETH if Treasuries pay more?

The base staking reward near 2.6 percent, or about 3.3 to 3.8 percent with MEV, now sits below a 10-year US Treasury near 4.8 percent. Stakers accept the gap because the reward is paid in ether, so the real return depends far more on the price of ETH than on the yield, and because treasuries and staking funds keep staking for reasons that are not sensitive to interest rates.

When is the Glamsterdam upgrade and will it make ETH scarce again?

Glamsterdam activated on the Sepolia test network on October 6, 2026, but the mainnet date slipped after a testnet activation was postponed over gas-repricing risks; developers are targeting early December, with December 1, 8, and 9 floated as candidates. It is a scaling upgrade that raises capacity, so it is more likely to suppress the fee burn than to revive it.

By Priya Reddy, senior markets writer at HOGE Wire.