AAVE Absorbed the Dilution. MORPHO Hasn't Started. SUSHI Shows Why It Matters.

Most investors model protocol revenue. Very few model who will own that revenue three years from now. SUSHI's annual revenue grew 21x from 2020 to 2021, and the token still collapsed 95%. The market dramatically underprices dilution and overpays for growth that won't accrue to existing holders. TI extends its token-vesting-supply framework with Ownership Decay , the velocity dimension that the static MC/FDV ratio misses.

1. The Pattern: Ownership Decay Is the Missing Metric

TI's token-vesting-supply framework uses the FDV-to-market-cap ratio as the quickest gauge of dilution exposure. It's a useful static screen: a ratio near 1.0 means most supply is circulating; a ratio of 3.0x or higher means the majority of supply has not yet entered the market. What it does not measure is velocity , how fast that locked supply will reach the market relative to demand growth.

This piece introduces Ownership Decay as the velocity dimension: the percentage of a holder's network ownership that will be diluted away over a fixed window. The formula is simple. For a 1-year window:

The reason this matters more than MC/FDV is that FDV is a snapshot of the destination. Ownership Decay is the speed of the journey. Two protocols with identical FDV ratios can have radically different decay rates if one front-loads its unlocks and the other back-loads them. And two protocols with identical revenue can produce radically different per-token value capture if one is issuing 2% of supply per year while the other is issuing 30%.

Read this chart and ask: are these five protocols being valued by the market in a way that reflects this dispersion? If you didn't know anything else about them, would you expect the token with ~1% annual decay to trade at the same earnings multiple as the token with ~16% annual decay and zero current revenue? The market currently prices MORPHO at $1.08B and AAVE at $943M. AAVE produces $125M of annualized revenue. MORPHO produces zero. The dilution gap is a 16x compounding mismatch. The valuation gap is roughly 13%, in the wrong direction.

2. AAVE: The Dilution Is Largely Behind It (with caveats)

AAVE has the cleanest supply story in major DeFi. 94.87% of the 16M maximum supply is circulating (15.18M tokens), market cap is $943M, fully diluted valuation is $994M, and the MC/FDV ratio is 0.95. Trailing twelve-month protocol revenue is $125M per DefiLlama. The April 2026 Aavenomics (AWW) proposal redirects 100% of protocol revenue from Aave-branded products to the DAO, eliminating application-layer value leakage that previously diluted the revenue accrual story.

The temptation is to describe AAVE as "equity-like." The metaphor is useful for thinking about the MC/FDV math , a token where almost all supply has reached the market behaves more like a static-share-count instrument than one with active emissions. But the metaphor is not a structural claim. AAVE token holders have no legal claim on cash flows, no court-enforced share register, no fiduciary duties owed to them by an issuer. The Safety Module continues to emit small but ongoing AAVE rewards as staking incentives. Governance retains discretion to expand supply if it chooses. Equity-like is the right shorthand for the supply mechanics; it is not the right framing for what kind of asset AAVE actually is.

With that caveat in place, AAVE's Ownership Decay over 1 year is approximately 1% , the Safety Module emission rate against the existing float. This is the lowest in the comparison set by an order of magnitude. The implication is direct: when AAVE moves on fundamentals, supply is not a confounding variable. Revenue growth, TVL expansion, or risk-management improvements flow through to existing holders without meaningful dilution friction. That is a structural advantage other DeFi tokens cannot claim.

It does not mean AAVE has zero supply risk. The current $943M market cap is down meaningfully from the levels at which the AWW program was approved, partly because the April 2026 rsETH/LayerZero contagion event re-rated Aave V3's pooled-liquidity architecture (TI's "Aave Concentrates, Kamino Isolates, Morpho Delegates" covers this in depth). The supply story is clean. The product-risk and architecture-transition story is not. The framework correctly separates the two.

3. MORPHO: The Dilution Is Enormous AND the Revenue Is Zero

This is the most undervalued risk in the comparison set, and the original auto-generated article missed the central fact entirely: MORPHO produces $0 of protocol revenue per DefiLlama. The fee switch is off. The protocol routes 100% of interest spread to lenders, with no take rate going to the DAO or token holders. There is no "future revenue accrual" thesis to anchor a valuation against, only the prospect that the fee switch turns on at some point and the protocol begins capturing value the way Aave and Compound do today.

Layer this against the supply schedule. MORPHO launched November 2024 with a 1B max supply, with core team, Morpho Labs, and strategic-partner allocations vesting over 36 months. Approximately 64.6% of supply is circulating today (645.96M tokens); the remaining 354M unlocks linearly through Q4 2027. That is approximately ~16% of current circulating supply entering the market every 12 months for the next two years, mostly to early backers and contributors.

The market cap is $1.08B and the FDV is $1.67B. The framework asks: what is the present value of a future stream of zero, divided by a denominator that will be 35% larger in 24 months? The honest answer is that MORPHO's valuation rests entirely on optionality , specifically the option that the fee switch turns on before the dilution pain lands. Every quarter that the switch stays off while team and investor tokens continue unlocking shifts more of the network's future value to people whose cost basis is zero.

This does not mean MORPHO is overvalued in absolute terms. The protocol has $6.24B of TVL per DefiLlama and a clear institutional adoption thesis. It does mean the per-token math is structurally adverse without a fee switch decision, and the market does not appear to be discounting that.

4. MKR/SKY: Supply Isn't Binary , The Sky Rebrand Proves It

The auto-generated article called MKR/SKY "capped + buybacks → deflationary." Post-Sky-rebrand, this is too simple. The actual mechanics are more interesting and the implications are different.

SKY total supply is 23.46B, of which 23.26B is circulating (99.1%). MKR-to-SKY conversion happens at 24,000:1 (one MKR redeems for 24,000 SKY), and the legacy MKR contract still has roughly 90,000 MKR uncoverted , equivalent to ~2.2B SKY, already counted in total supply. So the static MC/FDV picture is clean: $1.37B market cap against $1.38B FDV, ratio 0.99.

The dynamic picture is more complicated. Sky Token Rewards (STR) actively emit SKY to USDS holders. Per DefiLlama, the protocol has distributed $101M to holders over the trailing twelve months (the $102M+ figure the original article cited corresponds to this). At the current SKY price of ~$0.0589, that is approximately 1.71B SKY tokens issued through STR per year, against a 23.26B circulating base , an effective annual emission rate of ~7.3%.

STR emissions are matched, in part, by surplus burns funded by protocol revenue. Sky/Maker TTM revenue is $247M per DefiLlama, and the burn mechanism funded by surplus partially offsets the STR issuance. The net effect is not zero and not deeply negative; it is a redistributive flow from the protocol surplus through USDS yield back into SKY supply. The article's framing of "actively deflationary" describes the historical MKR Smart Burn Engine; it does not accurately describe the post-Sky-rebrand mechanism.

A more honest characterization: SKY is actively redistributive, not actively deflationary. The supply does not collapse over time; it cycles through the STR-USDS-surplus loop. For SKY holders who do not participate in STR, this means a ~6% net 1-year ownership decay against the static MC/FDV picture. For SKY holders who do participate (by holding USDS), the ownership-decay headwind reverses partially or fully. The protocol's design baselines redistribution; it does not baseline deflation.

This is the cleanest example in the comparison set of the expert critique that "supply isn't binary." A protocol can simultaneously run buybacks, fund holder distributions through token issuance, and net out to something that is neither pure inflation nor pure deflation. Calling it one or the other is what the market does, and what TI's framework should not.

5. AERO: Uncapped Emissions Mean Ownership Decays Continuously

The original article said AERO has a 1.89B total supply with no team unlock to 2028. The second half is correct in spirit but misleading in effect. The first half , the "1.89B total supply" framing , is the bigger problem, because it implies a cap that does not exist.

Verified data: AERO circulating supply is 947.9M, current total supply is 1.912B, and there is no maximum supply. AERO uses an emissions-based model similar to Velodrome's (its parent protocol), with weekly token emissions distributed to ve-AERO voters and LPs. The total supply grows continuously with each emission epoch. "1.89B total supply" is a snapshot of last week, not a destination.

The team-allocation cliff in 2028 is a real and meaningful detail. It means insider unlock pressure is deferred. But it does not mean the network is not being diluted in the interim , it is being diluted continuously through programmatic emissions to LPs and voters. AERO's 1-year ownership decay against the circulating float is approximately 27% on TI's reading of the last twelve months of issuance pace. That is the highest in the live comparison set, and it is the structural cost of the emissions-funded growth model.

The token is down 86.3% from its $2.32 ATH (December 2024) to its current $0.318. Part of that decline reflects the broader Solana-DeFi-summer compression that hit Jupiter, Kamino, and Raydium in the same window. Part reflects the AERO-specific cashflow drop TI's AERO research page documented in the May 2026 ERM investigation update , a 36% TTM decline in fee accrual to ve-holders. The ownership-decay story is the third layer: even if the cashflow recovers, ve-AERO holders who do not actively vote for high-emission pools are watching their network share erode by roughly 27% per year.

This is the case for an emissions-quality dimension to the framework, but it is also a structural argument against AERO at current levels for any holder who is not actively maximizing vote-bribe yield. The protocol's design says, explicitly: ownership goes to whoever shows up. Holders who do not show up lose share continuously.

6. The Dilution Graveyard: SUSHI 2021-22

The expert critique of the original article called for a "dilution graveyard" , a historical case where fundamentals improved and the token still collapsed because emissions overwhelmed value accrual. SUSHI is the canonical example, and the data is sharper than the suggestion implied.

From DefiLlama: SUSHI's annual protocol revenue grew from $4.09M in 2020 to $87.02M in 2021 , a 21x increase in a single year. By any standard fundamental-research metric, the protocol was executing. The token's all-time high was $23.38 on March 13 2021, set at the peak of the revenue surge. By the end of 2022, SUSHI had collapsed to under $1. Today it trades at $0.176, a 99.25% drawdown from its all-time high. Annual revenue had already decompressed back to $24.8M in 2022 and $6.7M in 2023, but the price decline outran the revenue decline by orders of magnitude.

The mechanism was supply. SUSHI's circulating supply grew from roughly 127M at the start of 2021 to roughly 250M at the start of 2022 , effectively doubling over the same 12 months that revenue grew 21x. SUSHI distributed those emissions to liquidity providers and stakers as a growth subsidy. The subsidy was real: it bought TVL, market share, and revenue. The cost was that existing holders' ownership share was nearly cut in half over the same window. The protocol grew. The per-token claim on that protocol collapsed.

This is the canonical proof that fundamentals analysis without supply analysis is incomplete. An investor in early 2021 looking at SUSHI's revenue trajectory had every reason to be bullish. The forecast that revenue would 20x in a year would have been right. The investor still lost ~95% over the following 24 months. The framework that would have flagged this risk is the one this report is extending: not "is the protocol growing?" but "is per-token value capture growing alongside protocol value creation?"

7. Where the Market Is Systematically Wrong

The contrarian read this report makes is not subtle: most crypto investors model revenue, and very few model who will own that revenue three years from now. The result is that the market systematically overpays for revenue growth in protocols that are diluting faster than they are growing, and systematically underpays for revenue clarity in protocols that have absorbed their dilution pain.

Three concrete mispricings this framework surfaces:

1. MORPHO and AAVE valuations are within 13% of each other. AAVE produces $125M of TTM revenue and has ~1% annual ownership decay. MORPHO produces $0 of TTM protocol revenue and has ~16% annual ownership decay. The market is pricing them as if their per-token value-accrual profiles were comparable. They are not.
2. AERO's emissions-funded growth model is being priced as if growth subsidizes ownership. It does not. Growth subsidizes the network in the abstract; ownership of that growth requires showing up at every vote epoch. Holders who treat AERO as a passive position are losing ~27% of their share per year.
3. SKY is being priced as "deflationary" by analysts who have not read the Sky rebrand carefully. It is redistributive, not deflationary, and the redistribution requires participation in USDS. Passive SKY holders are exposed to ~6% net annual decay; active participants are not. The framework should distinguish the two.

The simple version of the thesis: every protocol's revenue is being modeled with significantly more rigor than its supply. The arbitrage is small per-quarter and structural over years. SUSHI 2021-22 is not a curiosity; it is the cleanest worked example of the trade the market keeps repeating.

8. What TI's Framework Gains: Ownership Decay as the Velocity Dimension

This piece extends TI's existing Token Vesting & Supply concept page in one specific way: by adding Ownership Decay as the velocity counterpart to the existing MC/FDV ratio. The static ratio answers "how much supply has reached the market?" The velocity metric answers "how fast is more supply arriving relative to the float that already exists?" Both questions matter; the second one is the one most investors skip.

The four-input read this framework now produces:

Two observations. First, in any comparative valuation exercise across this set, the protocols with low decay should command structural premiums against the protocols with high decay, holding other variables constant. AAVE and SKY have that structural advantage; MORPHO and AERO have a structural disadvantage that the market is not fully pricing. Second, the framework correctly classified the SUSHI 2021-22 outcome ex-ante: a protocol with ~49% annual decay would have needed revenue to grow faster than the float was expanding just to keep per-token value capture flat, and even at +21x revenue growth, that did not happen. The lesson generalizes.

9. What Would Falsify the Framework

• AAVE falsifier: if AAVE underperforms BTC over the next two quarters while the buyback continues, the AWW revenue redirect maintains pace, and ownership decay stays near 1%, then the "supply clarity + product-protected revenue accrual" stack failed for reasons we have not named. Likely culprits: V4 transition risk, further pooled-liquidity contagion events, or a broader DeFi multiple compression that overwhelms the supply story.
• MORPHO falsifier (bullish): if MORPHO outperforms BTC over the next two quarters despite zero protocol revenue and 16% annual ownership decay, the framework owes an explanation. Most plausible mechanism: the market pricing in a near-term fee-switch decision; institutional vault adoption accelerating to the point that future revenue capture is being discounted aggressively.
• SKY falsifier: if passive SKY holders (who do not engage with USDS) outperform active SKY/USDS participants over two quarters, the "redistributive not deflationary" framing failed, and STR is functioning as a net burn rather than net issuance. Worth checking on the next monthly snapshot.
• AERO falsifier: if AERO recovers materially without a reduction in the emission rate, the model's prediction that passive ownership decays continuously needs a new dimension , likely that vote-bribe markets and protocol incentive efficiency make even passive ve-AERO holders better off than their decay rate implies.
• SUSHI generalizer: if no comparable historical pattern (revenue +N00%, token −N0%, supply +N00% in the same window) can be replicated from a different protocol cycle, the SUSHI case is a one-off curiosity rather than the canonical example. AVAX 2024 founder unlocks, OP/ARB 2024-25 cliffs, and CRV 2022-23 emissions all rhyme with the pattern; a more thorough historical sweep would strengthen the case.

10. The Honest Read

Revenue belongs to tokens. Future token count determines who receives it. The protocols that have already absorbed their dilution are increasingly behaving like businesses; the protocols still issuing large fractions of their supply are behaving like startups burning runway. The market mostly does not distinguish between the two. It models the revenue and ignores the denominator that will be drawn against that revenue when the time comes to value the network.

AAVE is closer to the business end of that spectrum than any other major DeFi token, with the structural caveat that "equity-like" is a metaphor and not a structural claim. SKY is closer than its post-rebrand mechanics make obvious, provided holders engage with the redistributive loop. MORPHO is at the maximum-dilution-with-zero-current-revenue extreme , the entire valuation rests on an optionality the market is not visibly pricing. AERO has built ownership decay into its core design as the cost of emissions-funded growth, which means passive holders are perpetually losing share.

SUSHI 2021-22 is the historical reminder that protocol growth and per-token value capture are different variables, and a research process that does not measure both will miss the trade. The auto-generator that produced the original draft of this article was building its conclusion on the wrong denominator; the rewritten framework includes the denominator explicitly.

The portfolio implication is direct: when comparing two tokens in this category, the question "how much of this network will I own three years from now?" deserves equal weight with "how much revenue will this network produce three years from now?" The first question is the one most investors are skipping. The arbitrage that fact creates is small per quarter and structural over years.