The Restaking Gold Rush Is Over: Why Ether.fi Is Walking Away From EigenLayer

For a stretch of 2024 and 2025, restaking was one of the hottest ideas in Ethereum. The pitch was simple: take ETH that is already staked, “re-use” it to secure additional services, and collect a second layer of yield on top of normal staking rewards. Billions of dollars poured in. Now, the data suggests that second layer of yield never really arrived, and one of the sector’s biggest players is walking away.

This piece is analysis and commentary on recently reported figures, not a forecast of where any token or protocol goes next.

What the numbers show

A CoinDesk analysis published on September 28 laid out the economics in stark terms. As of September 8, the restaking category held roughly $10 billion in deposits but generated only around $100,000 in fees over the prior week. Over the same period, plain liquid staking, with about $52 billion in deposits, produced roughly $27 million in fees. Per dollar secured, liquid staking was earning about 53 times more.

The protocol-level numbers tell the same story. According to that reporting, five of the largest liquid restaking protocols (Renzo, Kelp, Swell, Puffer Finance and Bedrock) produced combined gross profit of about $953,000 in the second quarter of 2026, down from about $2.18 million three quarters earlier. Some individual results were tiny relative to the capital raised to build them: Puffer reportedly booked around $21,600 in quarterly gross profit after raising $23 million in funding.

EigenLayer itself, now operating as EigenCloud, has seen deposits shrink from a peak above $20 billion to roughly $5 billion, according to the same analysis.

Ether.fi heads for the exit

The clearest signal came from ether.fi, which built its early growth by automatically routing depositors’ ETH into EigenLayer. Multiple outlets reported this week that ether.fi will cut its last structural ties to EigenLayer this quarter, leaving less than 1% of its assets restaked, and that it has already removed restaking from its weETH token. Its CEO, Mike Silagadze, framed the decision as one of risk versus reward: in his view, restaking still carried risk but no longer offered meaningful yield.

Ether.fi’s own financials show why it could afford to move on. Its gross profit reportedly fell about 47% from the third quarter of 2025 to the second quarter of 2026, but its card business generated more gross profit last quarter than restaking did. In other words, restaking had become a secondary line rather than the core of the business.

Why the yield never showed up

Our read is that restaking ran into a basic supply-and-demand problem. Restaking creates a supply of “shared security” that other networks and services (often called actively validated services, or AVSs) can rent. For that to pay depositors a premium, those services have to pay real fees, and enough of them to cover both base staking rewards and the extra risk. Based on the reported fee data, that demand side simply has not grown into the supply that was built.

Two other shifts made the gap more obvious:

• Points programs ended. Much of the early growth was driven by points and airdrop expectations rather than organic fees. Those subsidies wound down through 2025.
• Slashing went live. Slashing on EigenLayer was switched on in April 2025, turning what had been a theoretical downside into a real, priced one. Taking on more risk for a shrinking premium is a hard sell.

The sector also carries scars from security incidents. CoinDesk’s analysis highlighted an April exploit involving Kelp’s bridge, in which an attacker reportedly minted a large amount of unbacked rsETH, with knock-on effects for lending markets that accepted the token as collateral. Incidents like that reinforce the point that stacking protocols on top of each other stacks risk too.

Where the capital is going instead

Restaking is not disappearing entirely. Ether.fi reportedly moved the restaking it still does to Symbiotic, a rival platform that says it has integrated more than 50 networks. But a meaningful share of capital appears to be rotating toward more conventional yield sources, such as curated lending vaults. Morpho’s curated vaults, for example, were cited as holding around $5.8 billion.

That shift fits a broader pattern in crypto this year: yield products increasingly have to be backed by identifiable cash flows, such as borrowing demand, trading fees or real-world assets, rather than token incentives and future promises.

What it means for everyday ETH holders

For readers who hold liquid staking or liquid restaking tokens, a few practical takeaways stand out:

• Check what your token actually does now. Products change underneath users. If you hold weETH or another liquid restaking token, read the protocol’s latest documentation to understand whether your ETH is still restaked, and where.
• Compare risk, not just headline APY. An extra layer of protocol and slashing risk only makes sense if it is paid for. The reported fee data suggests that, for now, it often is not.
• Watch collateral chains. When a token is used as collateral elsewhere in DeFi, a problem in one protocol can spread. Understanding those links matters more than chasing a fraction of a percent in yield.

The bigger lesson

Restaking was a genuinely novel idea, and it may still find a sustainable niche if demand for shared security grows. But its first cycle looks like a familiar crypto arc: a compelling narrative, heavy incentives, rapid inflows, and then a reckoning once the incentives fade and the real fee base is measured. The protocols that survive are likely to be the ones that find revenue beyond the narrative, as ether.fi appears to be attempting with its card business.

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