The Yield Illusion: Where Does Restaking APY Actually Come From?

[defi_diana]

Following the Money in a $16B Ecosystem

If you’ve been in DeFi for more than one cycle, you know to ask one question before depositing capital anywhere: where does the yield come from? When the answer is vague, hand-wavy, or involves the phrase “sustainable tokenomics,” that’s your signal to run.

With EigenLayer/EigenCloud holding $16B TVL and restaking protocols advertising yields from 5% to 15%+, I’ve spent the last month trying to trace exactly where restaking APY originates. The results are… enlightening.

Decomposing Restaking Yield

Let me break down the actual sources of restaking yield, layer by layer:

Source 1: Base Ethereum Staking Yield (~3.2-3.8%)

This is the only truly organic and sustainable component. When you restake ETH (either natively or via an LST), you continue earning Ethereum’s base staking yield from:

• Block proposals (execution layer rewards)
• Attestation rewards (consensus layer)
• MEV tips (via MEV-Boost)

This yield is real, well-understood, and has been consistent for years. It’s also available to any ETH staker without restaking. Restaking doesn’t increase this base yield — it’s just the floor.

Source 2: AVS Fee Revenue (~0.1-0.5% estimated)

This is what restaking was supposed to be about. AVSs pay fees to operators/restakers in exchange for economic security. In theory, as more AVSs launch and gain traction, this yield component grows.

In practice, AVS fee revenue is negligible. I’ve tracked on-chain fee flows from the major AVSs and the total annualized fee revenue across all AVSs is in the low millions — against $16B in restaked capital, that’s approximately 0.01-0.05% additional yield.

Even being generous and assuming AVS revenue grows 10x in the next year, we’re still talking about less than 0.5% additional yield. This is the honest answer to “what do AVSs pay?” — barely anything, at least right now.

Source 3: EIGEN Token Emissions (~1-4%)

Here’s where things get murkier. EigenLayer distributes EIGEN tokens as rewards to restakers and operators. The value of these emissions depends on EIGEN’s market price, which has dropped ~87% from its all-time high.

At current prices, EIGEN emissions add roughly 1-2% to restaking yields. But this is dilutive — each new EIGEN distributed dilutes existing holders. It’s the same mechanic as farm-and-dump yield farming from 2020-2021, just with fancier branding.

Key insight: EIGEN emissions are not yield. They’re dilution of existing token holders subsidizing restaker returns. When EIGEN’s price drops (as it has been), this “yield” evaporates.

Source 4: Points Programs (~??? to ???%)

The most opaque yield source of all. Multiple LRT protocols (EtherFi, Renzo, Kelp, Puffer) ran or continue to run points programs that reward depositors with protocol-specific points, which may or may not convert to tokens in future airdrops.

The “yield” from points is impossible to calculate in advance because:

• The token valuation is unknown until airdrop
• The conversion ratio from points to tokens is unknown
• The amount of future dilution is unknown
• There’s no guarantee of any airdrop at all

Despite this, points programs were the primary driver of restaking TVL growth during 2024-2025. Many depositors were earning “yield” that existed entirely as spreadsheet projections of potential airdrop values.

Source 5: LRT Protocol Incentives (~0.5-3%)

LRT protocols themselves often subsidize yields to attract deposits. This can take the form of:

• Protocol token rewards (ETHFI, REZ, etc.)
• Boosted yields for specific pools
• Partnership incentives

Like EIGEN emissions, these are dilutive and unsustainable. They work to bootstrap TVL but create a dependency on continued token emissions to maintain yields.

The Yield Decomposition Table

Source	Estimated APY	Sustainability	Source of Value
Base ETH staking	3.2-3.8%	High	Ethereum consensus + MEV
AVS fees	0.01-0.5%	Could grow	Service demand
EIGEN emissions	1-4%	Low	Token dilution
Points/airdrops	Unknown	Very low	Speculation
LRT incentives	0.5-3%	Low	Protocol token dilution
Total advertised	5-15%+	Mostly low	Mostly unsustainable

Why This Matters

The honest yield from restaking is approximately 3.5% — which is just the base Ethereum staking rate. Everything on top of that is either:

1. Negligible (AVS fees)
2. Token dilution (EIGEN/LRT emissions)
3. Speculation (points/airdrops)

This doesn’t mean restaking yields will immediately collapse. Token emission programs can last for years, and new LRT protocols launching keep the incentive cycle going. But it does mean that anyone earning “10% on restaked ETH” should understand that 6-7% of that yield is coming from sources that will eventually end.

The Comparison to Luna/Anchor

I know this comparison triggers people, but hear me out. Anchor Protocol offered ~20% on UST, and the yield came primarily from:

1. A small amount of real lending revenue
2. Massive Anchor Reserve subsidies (effectively, emissions)

Restaking yields are structurally similar:

1. A small amount of real AVS revenue
2. Massive EIGEN/LRT token subsidies

The scale is different (20% vs 5-15%), and restaking doesn’t have the algorithmic stablecoin risk. But the yield-source opacity is uncomfortably similar.

What Needs to Change

1. Yield source labeling: Every restaking protocol should clearly decompose its advertised yield into organic (AVS fees) and inorganic (emissions) components
2. AVS revenue dashboards: Real-time tracking of actual AVS fee payments, so users can see the organic yield
3. Emissions schedule transparency: Clear documentation of when EIGEN/LRT emissions will taper and what yield looks like without them
4. Points program honesty: Stop advertising speculative airdrop values as “yield”

The restaking ecosystem has real potential. But we can’t build sustainable infrastructure on yield illusions. The sooner we’re honest about where the APY comes from, the sooner we can focus on growing the organic revenue that actually justifies the $16B in restaked capital.

If you’re currently restaking, have you calculated your yield decomposition? What percentage is organic vs emission-based?

---

Sources: DatoWallet Ethereum Staking Statistics, CoinLaw Liquid Staking Adoption, QuickNode Restaking Revolution, ArXiv LST Restaking Trends

[crypto_chris]

Diana, this yield decomposition should be required reading for anyone in restaking. I’ve been saying this for months but you’ve laid it out more clearly than anyone.

I want to add one more yield source that’s often overlooked: opportunity cost arbitrage.

Many institutional restakers aren’t actually looking at the absolute yield. They’re comparing restaking yield to the alternative — holding ETH in a cold wallet with 0% yield, or staking directly for ~3.5%. Since restaking adds some additional yield (even if mostly from emissions), the marginal decision for an already-staked ETH holder is: “do I earn an extra 1-3% for accepting additional slashing risk?”

Framed this way, the calculation changes. It’s not “is 5% sustainable?” — it’s “is the extra 1-3% above base staking yield worth the additional risk?” And for many institutions with large ETH holdings, the answer is yes, even if that extra yield comes from emissions, because:

1. The slashing risk has been zero so far (no major slashing events)
2. The opportunity cost of NOT restaking is leaving yield on the table
3. The institutional FOMO of underperforming peers who are restaking

This is exactly how mispriced risk builds up in financial systems. Everyone knows the extra yield is subsidized. Everyone knows slashing could happen. But nobody wants to be the one who opted out while their competitors earned an extra 2%.

It’s a collective action problem. The individually rational decision (restake for the marginal yield) creates systemic risk at scale. We saw this exact dynamic in the pre-2008 CDO market — every individual participant was making a rational decision, but the system as a whole was accumulating risk that nobody was pricing correctly.

The Luna/Anchor comparison is valid but incomplete. The key difference is that restaking doesn’t have a reflexive death spiral mechanism. There’s no algorithmic peg to defend. The worst case for restaking is a gradual TVL decline as emissions taper, not a sudden collapse. Unless there’s a slashing event, in which case all bets are off.

[data_engineer_mike]

I’ve actually built dashboards tracking restaking yield sources on-chain, so let me share some real data:

AVS fee revenue tracking (last 90 days):

I indexed all reward distribution events from the EigenLayer reward contracts. Here’s what I found:

• Total AVS fee distributions to operators/restakers: approximately $4.2M over 90 days
• Annualized: ~$16.8M
• Against $16B in restaked capital: 0.105% annualized yield from organic AVS fees

That number is actually lower than Diana’s estimate. And it’s concentrated — EigenDA accounts for roughly 60% of all AVS fee revenue. The remaining 40+ AVSs collectively generate less than a single medium-sized DeFi protocol’s fee revenue.

EIGEN emission tracking:

• Over the same 90-day period, ~$180M worth of EIGEN was distributed (at time of distribution)
• Much of this EIGEN was immediately sold by recipients
• The selling pressure from EIGEN emissions is a contributing factor to the 87% price decline

Here’s the chart that tells the whole story: if you plot EIGEN emissions over time against EIGEN price, there’s a clear inverse correlation. More emissions = more selling = lower price = lower “yield” from emissions = need for more emissions to maintain the same dollar-denominated yield. It’s a dilution spiral.

Points valuations:

I tracked the major LRT airdrop outcomes:

• EtherFi (ETHFI): initial airdrop valued at ~3-4% of deposited capital, now trading at ~1.5% annualized value
• Renzo (REZ): initial airdrop valued at ~2-3%, with significant price decay post-launch
• Other LRT airdrops have been similarly disappointing relative to projections

The gap between “expected points value” and “realized airdrop value” is consistently 50-70%. People who deposited based on optimistic point projections systematically overestimated their returns.

My recommendation: I built an open-source dashboard that decomposes restaking yields in real-time. It categorizes every reward distribution as either organic (AVS fees) or inorganic (emissions). If anyone is interested, I’ll share the link. The industry needs tools like this to be standard, not side projects.

[product_manager_alex]

This thread perfectly illustrates why I think the restaking ecosystem has a narrative problem, not (just) a yield problem.

The original EigenLayer thesis was: “build a marketplace for decentralized trust.” AVSs need security, restakers provide it, the marketplace matches supply and demand, and yield reflects the price of security. Beautiful, elegant, and completely different from what actually happened.

What actually happened:

1. Points meta drove TVL to $20B on the supply side (restakers depositing for points)
2. Almost nothing happened on the demand side (AVSs paying for security)
3. The protocol responded by launching EIGEN tokens to maintain the illusion of yield
4. The rebrand to EigenCloud is an attempt to pivot the narrative toward a broader story

This is a product-market fit problem. EigenLayer built a brilliant supply-side product (“restake your ETH for more yield”) but the demand side (“pay for restaked security”) hasn’t materialized. This is exactly the situation you’d expect to produce the yield decomposition Diana described — lots of supply-side incentives masking weak demand.

The comparison I’d make isn’t Luna/Anchor. It’s early Uber. Uber subsidized rides for years to build supply (drivers) and demand (riders), running massive losses. The question was always: when subsidies end, does the marketplace stand on its own? For Uber, the answer was eventually yes (barely). For restaking, the jury is still out.

What would “product-market fit” look like for AVSs?

• 10+ AVSs generating >M/year each in organic fees
• Total AVS revenue sufficient to provide ~1%+ yield on restaked capital without emissions
• Growing demand from new AVSs willing to pay meaningful fees
• Reduced dependency on EIGEN emissions to attract restakers

We’re nowhere near these benchmarks today. That doesn’t mean we won’t get there — marketplace businesses are notoriously slow to reach equilibrium. But at $16B TVL, the expectation bar is set very high.

Mike — please share that dashboard. The industry desperately needs transparent yield attribution tools.