Liquid Staking Protocols Now Hold More ETH Than the Beacon Chain Deposit Contract

A milestone quietly passed that should make every Ethereum enthusiast both proud and nervous.

Liquid staking protocols now collectively hold more ETH than what individual validators have deposited directly through the Beacon Chain deposit contract. Lido alone sits at $34 billion in TVL. Add Coinbase's cbETH, Binance's staked ETH ($11 billion), Rocket Pool, ether.fi ($10 billion), and the rest, and the total liquid staking TVL dwarfs direct staking.

This is a triumph of product-market fit. It's also a concentration risk that Ethereum's architects never intended.

How We Got Here

The problem liquid staking solves is real and obvious.

To stake ETH directly, you need 32 ETH (roughly $63,000 at current prices). You need to run a validator node 24/7. You need technical knowledge to maintain it. And until relatively recently, your ETH was locked with no withdrawal ability.

Liquid staking protocols removed every one of those barriers. Deposit any amount of ETH. Receive a liquid token (stETH, cbETH, rETH) that represents your staked position. Use that token in DeFi. Earn staking rewards without running hardware.

The product is so obviously better than direct staking for most people that adoption was inevitable. What wasn't inevitable was the degree of concentration.

The Lido Question

Lido holds roughly 28-29% of all staked ETH. That's a staggering concentration for a single protocol.

Why does this matter? Because Ethereum's security model assumes that no single entity controls more than 33% of stake. At 33%, an entity can prevent finality. At 50%, it can censor transactions. At 67%, it can rewrite history.

Lido isn't at 33% yet, and the Lido DAO has committed to self-limiting. But "committed to self-limiting" is a governance decision, not a protocol constraint. If ETH depositors keep choosing Lido over alternatives, that commitment gets tested.

Lido's dominance also creates a weird dynamic. stETH has become the de facto standard for staked ETH in DeFi. Most lending protocols accept stETH as collateral. Most yield strategies use stETH as the base layer. This network effect makes it harder for competitors to catch up, reinforcing the concentration.

The Centralized Staking Elephant

Here's the part that gets less attention than Lido's dominance: centralized exchanges collectively control an even more concerning share.

Coinbase's cbETH, Binance's staked ETH, and Kraken's staking service (before regulatory action) combined hold a massive percentage of staked ETH. Binance alone has $11 billion worth.

The difference between Lido and Coinbase is important. Lido runs through a set of vetted but independent node operators. If the Lido DAO makes a bad decision, node operators can theoretically resist. Coinbase runs its own validators. If the US government orders Coinbase to censor certain transactions at the validator level, Coinbase has both the technical capability and the legal obligation to comply.

We saw a preview of this with OFAC sanctions and Tornado Cash. When the US government sanctioned Tornado Cash addresses, the question became: would validators censor transactions to and from those addresses? Coinbase's CEO said they would comply with government orders. That means a significant chunk of Ethereum's staked ETH is operated by entities that will censor transactions if ordered to.

The DeFi Composability Layer

Liquid staking tokens have become critical infrastructure for all of DeFi. This is both their greatest success and their biggest systemic risk.

stETH is used as collateral on Aave, Maker, and dozens of lending protocols. It trades on every major DEX. It's the base asset for restaking protocols like EigenLayer ($18 billion TVL). It's wrapped, rehypothecated, and layered across the entire DeFi stack.

If stETH depegs from ETH, even temporarily, the cascading effects would be enormous. Liquidations across lending protocols. Panic selling on DEXs. Restaking positions unwinding. We got a small taste of this in June 2022 when stETH briefly depegged during the Three Arrows Capital collapse.

The depeg risk is theoretically small now that withdrawals are enabled (stETH can always be redeemed for ETH, just with a time delay). But during a crisis, "just wait a few days for your withdrawal" isn't comforting when your collateral is being liquidated in real-time.

The Solo Staker Squeeze

Solo staking is dying. Not because it doesn't work, but because it can't compete.

A solo staker earns roughly the same yield as a Lido depositor, but they need 32 ETH, technical expertise, and dedicated hardware. They can't use their staked ETH in DeFi. They bear all the operational risk of running a node.

A Lido depositor earns yield, keeps liquidity, can use stETH across DeFi, and doesn't need any technical knowledge. They pay a 10% fee to Lido, but the convenience premium is worth it for most people.

The result is a steady migration from solo staking to liquid staking. Every quarter, the percentage of staked ETH in liquid staking protocols grows, and the percentage in solo validators shrinks.

Ethereum was designed for a world of many independent validators. We're building a world of a few large staking providers. Those providers operate multiple validators, but the control is concentrated. And that's a fundamental deviation from the original vision.

What Can Be Done

Several approaches are being discussed:

Protocol-level limits. Could Ethereum enforce a cap on how much ETH any single entity can stake? Technically possible but practically difficult. Lido could split into multiple entities. Coinbase could create subsidiary validators. Concentration is hard to prevent at the protocol level.

Distributed validator technology (DVT). Obol Network and SSV Network are building DVT, which lets multiple operators collectively run a single validator. This reduces the risk of any single operator going down or being coerced. Lido is exploring DVT integration, which could mitigate some centralization concerns even while Lido's market share stays high.

Economic incentives. Some proposals would reduce staking rewards when the total stake ratio gets too high, naturally discouraging marginal stakers and reducing the total amount of staked ETH.

Education and tooling. Making solo staking easier (lower deposits, better software, simpler setup) could slow the migration to liquid staking. But this is fighting against powerful economic forces.

My Take

Liquid staking's dominance is a feature, not a bug, from a user perspective. The product is better. People are making rational decisions to use it. You can't fault individuals for choosing Lido over running their own validator.

But from a network health perspective, it's concerning. Ethereum's decentralization and censorship resistance depend on a diverse validator set. When a handful of protocols and exchanges control most of the stake, those properties are at risk.

The good news is that the Ethereum community is aware of the problem and actively working on solutions. DVT, better solo staking tools, and protocol-level adjustments could help.

The bad news is that market forces are powerful, and the trend toward concentration is accelerating. Solving this requires either making solo staking competitive with liquid staking (hard) or making liquid staking more decentralized (possible but slow).

This is one of Ethereum's most important challenges. And unlike scaling or fees, there's no clear technical solution. It's a coordination problem, and those are always the hardest to solve.