Ethereum is trading around $1,800 tonight, down 25% in a month and 28% in two weeks. The ETH/BTC exchange rate has slipped to 0.029, approaching a three-month low. Bitcoin is down about 11% year to date, while Ethereum is down about 32%. Where did those extra 20 percentage points go? Let’s break down the numbers today.

The first problem: the burn engine has stalled.

EIP-1559 was designed to burn the mainnet base fee directly. Back in 2021–2023, the “ultrasound money” narrative depended on burns outpacing issuance. But that’s no longer happening: so far in 2026, the entire network has burned only about 19,000 $ETH, compared with 1.48 million for all of 2022. That’s less than 3% of new issuance. Supply rose from 121.34 million at the start of the year to 122.2 million on November 8, a net increase of about 865,000 so far this year, or an annualized inflation rate of about 0.7%. The reason is that activity has moved to L2s: mainnet transactions averaged 2.17 million per day this year, up 54% from last year, while fees fell by 58. A median transfer now costs just 3.5 cents. Since the Dencun upgrade, blob space has burned only about 4,700 ETH in total—the roadmap was deliberately designed to make it cheap, at the cost of leaving the mainnet with little revenue.

Second issue: value capture has moved elsewhere.

The Q2 data compiled by on-chain analyst Tanaka paints a clear picture: applications in the Ethereum ecosystem generated $1.79 billion in fees, while the mainnet captured just 4.9% of that, or about $88.4 million. L2s process around 23.3 million transactions a day—ten times the mainnet’s volume. Users live on rollups, Base and Robinhood Chain control the on-ramps, and the settlement layer is left with the rent. Standard Chartered put it more bluntly: L2s are extracting excess profits from the Ethereum ecosystem.

Third issue: the biggest spot buyer has pulled back.

BitMine holds 6.06 million ETH, nearly 5% of the circulating supply, at an average cost of about $3,500, with over $4 billion in unrealized losses. Its management did the math: its stock was trading at just 0.88 times net asset value, so last week it spent $45.9 million buying back shares and only $24.5 million buying crypto. Buybacks have totaled nearly $400 million this year. ETFs are also pulling back: U.S. spot Ethereum ETFs saw net outflows of more than $1.2 billion in October, led by BlackRock’s ETHA. In the first ten months of this year, only two months saw net inflows. And ETFs can’t stake, so their holdings earn no yield.

But on-chain, the picture looks different: about 1.37 million ETH are queued for staking activation, with a 24-day wait; the exit queue has only 13,000 ETH, which would clear in just over a day. The share of ETH staked rose from 27.6% to 36.1% over the past year. Vitalik has also weighed in: ETH is a store of value, not a fee-burning token, and it’s fine if less ETH is burned.

The criteria can actually be quantified. Just watch three signals: first, whether the mainnet base fee can rise from around 0.3 gwei back to about 13.6 gwei—the level at which burns offset issuance; second, whether L2s pay a genuine scarcity price for blob space; and third, whether roughly $299.4 billion in stablecoins and $17.2 billion in tokenized assets on-chain are actually being used, rather than sitting idle.

For your portfolio, this means ETH’s valuation anchor is shifting from “fee revenue” to “reserve asset of the settlement layer.” Until the shift is confirmed by data, its volatility will be driven mainly by capital flows, with little relation to how well the network works. This is not a call on whether the price will rise or fall; it comes down to one thing—whether you can withstand this level of volatility.

Do you think ETH can ultimately shift its value anchor successfully, or will it never return to the fee narrative?

This article is for informational and educational purposes only and does not constitute investment advice.

Sources: ultrasound.money, Ethereum Institutional (State of the Ledger) (November 5), Tanaka_L2’s Q2 value capture analysis, BitMine’s 10-K and November 9 announcement, Farside Investors, The Block, CoinDesk.