Here's a number most dividend investors have never seen: in the months following Ethereum's EIP-1559 upgrade and before the mass adoption of data blobs, the network was destroying more ETH than validators minted—supply was actively shrinking. The mechanism that makes this possible is still running. Right now, as of March 2026, Ethereum issues roughly 2,800 ETH per day to stakers and burns roughly 2,300 ETH per day in fees, leaving it barely inflationary at under 1% annually. That's a far cry from the ~4% annual inflation of fiat currencies, and it sits on a hair trigger: any sustained spike in on-chain activity could tip it deflationary again overnight.

A company returning capital to shareholders at a comparable rate via buybacks would have financial media running columns about capital discipline and shareholder-friendly management.

Instead, it happens quietly inside a blockchain. And almost no income-focused investor notices the mechanism at all.

That's a mistake worth correcting—because ETH in 2026 is not the same speculative token it was in 2021. The plumbing changed. The economics changed. And for a traditional portfolio built around dividend yield and compounding, there's now a legitimate case for a small allocation that goes beyond "crypto is just speculation."

The Engine Room: EIP-1559 and the Burn That Never Stops

To understand why ETH supply is shrinking, you need to understand one upgrade: EIP-1559, which went live in August 2021. Before that, Ethereum worked like most blockchains—miners competed for transaction fees, and all those fees flowed to them. The more congested the network, the more miners earned.

EIP-1559 broke that model in a specific way. It split every transaction fee into two pieces:

• A base fee, which is algorithmically set and sent directly to a burn address—destroyed permanently, removed from supply forever.
• A tip (or priority fee), paid directly to validators as an incentive to include the transaction quickly.

The base fee is the important part. Every single time someone swaps tokens, buys an NFT, interacts with a DeFi protocol, or settles a Layer-2 batch on Ethereum's mainnet, a portion of the fee they pay ceases to exist. No wallet receives it. It's gone.

Why Layer-2 Activity and the Burn Rate Have a Complicated Relationship

Here's where it gets counterintuitive, and where a lot of crypto commentary gets it wrong.

Layer-2 networks like Arbitrum, Base, Optimism, and zkSync execute thousands of transactions cheaply off Ethereum's main chain. They periodically bundle those transactions and "settle" them back onto Ethereum's mainnet—historically posting compressed transaction data that paid a mainnet fee, part of which got burned.

But in March 2024, Ethereum implemented EIP-4844 (part of the Dencun upgrade), which introduced a new, cheaper data type specifically for L2 settlement called blobs. Blobs have their own separate fee market and are far cheaper than the old calldata method. This was great for L2 users—transaction fees on Base and Arbitrum dropped by 90%+. For ETH's burn rate, it was a structural cut: the fee revenue from L2 settlements collapsed, and with it, the deflationary pressure those settlements used to generate.

This is why you can see a paradox in the data today: L2 networks like Base process millions of transactions daily—more than many traditional payment networks—yet the ETH burn rate from L2 settlement is a rounding error compared to what it was in 2023.

The burn engine now runs primarily on direct mainnet activity: DeFi protocols, stablecoin transfers, high-frequency trading bots, and NFT markets. When those heat up, the burn rate spikes and can briefly exceed issuance. When they cool down—as they have in early 2026—the network tips slightly inflationary.

The threshold matters: Ethereum needs roughly 2,800+ ETH burned per day to offset new issuance at current staking levels. At moderate network activity, it comes close. At peak DeFi seasons, it crosses it comfortably. That knife-edge is the point.

ETH Staking: The Closest Thing to a Dividend in Crypto

Here's where income investors should pay close attention.

When Ethereum switched from Proof-of-Work to Proof-of-Stake in September 2022 (the event called "The Merge"), it replaced energy-intensive mining with a system where validators lock up ETH as collateral to secure the network. In return, they earn newly issued ETH and a portion of transaction tips. Currently, staking ETH yields approximately 2.8–3.2% annually, depending on network conditions and whether you stake directly or through a liquid staking protocol—with the Staking Rewards Benchmark Rate sitting at roughly 3.0% as of early 2026.

That yield is not theoretical. It pays out in ETH, continuously, as blocks are produced roughly every 12 seconds.

Compare that to a traditional dividend portfolio:

• The average S&P 500 dividend yield: ~1.4%
• 10-year Treasury note: ~4.3%
• Staking ETH: ~2.8–3.2%

The staking yield alone puts ETH in the same conversation as many investment-grade bonds—except the underlying asset has, historically, appreciated in price during bull cycles in ways bonds simply do not.

How Investors Actually Access Staking Yield

You don't need to run a validator node (that requires 32 ETH and technical setup). There are cleaner paths:

Liquid staking protocols like Lido issue you stETH—a token representing your staked ETH plus accruing rewards. You can hold it, trade it, or use it in DeFi. The yield accrues automatically.

Spot Ethereum ETFs in the U.S. currently hold ETH but generally do not pass staking yield to shareholders due to regulatory structure. BlackRock's iShares Ethereum Trust is the major example:

ETHA gives you price exposure without the staking yield—think of it like owning gold through GLD: you get the commodity price movement, but no carry income. If you want the yield, you need to hold ETH directly or through a yield-passing wrapper in a non-U.S. jurisdiction that permits it.

This distinction matters enormously for an income investor. ETHA is a price bet. Staking ETH through a liquid protocol is an income position.

The Buyback Analogy That Actually Holds Up

Traditional finance has a concept income investors understand well: share buybacks reduce outstanding shares, and each remaining share represents a larger slice of the same economic pie. When a company destroys its own equity, existing shareholders get richer in proportional terms even without a cash dividend.

ETH's burn mechanism is structurally identical—except the "company" is an open financial network, and the "shareholders" are everyone who holds ETH.

Microsoft buying back shares doesn't mean every shareholder receives a check. But the value of each remaining share increases. Same principle. The difference is that Ethereum's "buyback program" is executed automatically by market demand—the more people use the network, the faster supply shrinks.

What This Means for a Traditional Income Portfolio

None of this makes ETH a replacement for your dividend stocks. A 4% staking yield on an asset that can drop 40% in a quarter is not the same as a 4% yield on a utility stock that's traded in a tight range for a decade.

But that volatility argument cuts both ways. The low correlation is the point.

From 2022 through 2025, Ethereum's rolling 90-day correlation with the S&P 500 averaged roughly 0.35—meaningfully lower than, say, small-cap growth stocks. During specific stress periods (March 2020, the 2022 rate-shock selloff), ETH diverged sharply from equity markets in both directions.

A hypothetical portfolio constructed around this logic: imagine a traditional income investor—call her Sarah, 48, with a $500,000 dividend portfolio heavily weighted toward Realty Income, JPMorgan, and dividend ETFs. She allocates 3% ($15,000) to ETH via a liquid staking protocol at 3.5% yield. Her ETH position generates roughly $525 in staking rewards annually. That alone won't retire her. But it does something her bond allocation can't: it participates in potential price appreciation while paying carry, and it moves differently from the rest of her portfolio when markets get weird.

The sizing matters as much as the asset. At 1–3%, ETH adds diversification and modest yield. At 15%, it shifts the entire risk profile of the portfolio into territory that income investors should avoid.

The Red Flags Income Investors Should Not Ignore

Being intellectually honest about this requires acknowledging what can go wrong:

• Regulatory risk is real. The SEC's treatment of ETH as a commodity versus security has shifted multiple times. A hostile regulatory ruling could restrict access to staking yield for U.S. investors through mainstream platforms.
• Protocol risk. Ethereum has executed major upgrades successfully, but smart contracts fail. Liquid staking protocols like Lido have had minor incidents and carry counterparty risk that a T-bill does not.
• Tax treatment is a mess. Staking rewards are taxed as ordinary income in the U.S. at the time of receipt, not at the lower qualified dividend rate. For high earners, this meaningfully erodes the yield advantage.
• Slashing risk. If validators behave maliciously or experience downtime, they can "slash"—lose a portion of their staked ETH. Liquid staking protocols socialize this risk across all participants, but it doesn't disappear.

What to Watch in the Next 7 Days

The metric that will tell you whether ETH tips deflationary—or stays inflationary—is brutally simple: the daily burn rate versus daily issuance. You can track it in real time at ultrasound.money.

The post-EIP-4844 reality is that L2 settlement no longer drives the burn meaningfully—blobs are too cheap. The burn now depends on direct mainnet activity: DeFi volumes, stablecoin flows, MEV activity, and high-frequency contract interactions. If those stay suppressed, ETH stays slightly inflationary. If they reaccelerate, the network can tip deflationary within days.

One specific thing to watch: Ethereum mainnet gas prices. Check etherscan.io/gastracker. When the base fee sustains above roughly 15–20 Gwei, Ethereum is burning faster than it's issuing. When it sits at the current 0.1–1 Gwei range seen in early 2026, it isn't. That single number tells you more about the ETH supply narrative than any pundit will.

Does that make ETH an income asset? Or is it still just a high-volatility growth bet wearing a yield costume?