Protocol Capital Structure: Equity, Debt, and Whatever Onchain Protocols Are Building Instead
A credit agency finally scored one. The answer came back at 0.9%, and it says more about onchain finance than any TVL chart.
Corporate finance hands you two boxes. Equity and debt. Onchain protocols do not fit in either.
On 7 October 2026, Moody’s tried anyway. It assigned Sky Protocol a B3 issuer rating with a stable outlook, the first rating it has ever issued to a stablecoin protocol. S&P Global had already landed on B-.
Two agencies. Two methods. One shared conclusion: this thing has a balance sheet now, and the capital sitting under it is thin.
I have spent the last few weeks reading protocol financials properly. The capital structure question is the one almost nobody answers.
A protocol does not issue shares or file a bond prospectus. It still has creditors, a first-loss layer and a residual claim. They just live in smart contracts.
What is a protocol capital structure, and why is everyone suddenly asking?
Answer first. A protocol capital structure is the order in which claims on an onchain balance sheet get paid, and the order in which they absorb losses.
Same job as a corporate capital stack. Completely different instruments.
- Senior claims sit at the top. For USDS, that is whoever holds the stablecoin.
- Loss-absorbing capital sits underneath. Reserves, buffers, retained surplus.
- The residual claim sits at the bottom. That is the governance token.
- Nobody signs an indenture. The waterfall is code, plus a vote.
So why now? Because the numbers finally exist in a form an analyst can read.
Sky Frontier Foundation publishes a quarterly balance sheet for Sky Protocol. Credit teams picked it up. Then they graded it.
The timing is not an accident either. Stablecoins crossed $300B in supply this cycle. At that size, “trust me, it is overcollateralised” stops being an answer.
Where does the debt sit when a protocol has never issued a bond?
In the stablecoin itself.
USDS is the liability. Every unit is a dollar claim the protocol has to honour on demand.
The other side of the ledger is collateral. At the time of writing, $14.15B of Protocol Collateral against $11.48B of stablecoin supply.
That gap is the entire argument.
- Q2 2026 closed with $12.32B of collateral against $12.22B of obligations.
- The $90.26M difference is Protocol Surplus. Equity in everything but name.
- Protocol Collateral grew 45.5% year over year.
- USDS is not borrowed from a lender. It is drawn against approved collateral, or converted 1:1 through the Peg Stability Module.
What counts as equity when a protocol cannot issue shares?
Three layers, stacked.
- Protocol Surplus. Retained earnings under a different name.
- Sky Reserves. The dedicated loss buffer, at $80.36M on 8 October against a governance-set $150M solvency target.
- SKY. The residual claim. 17.08B staked at the Q2 close, or 73.3% of circulating supply.
Now the uncomfortable detail. Sky Reserves went down.
They stood at $93.53M at the end of August. Six weeks later, $80.36M.
A reserve that only ever rises is a marketing number. This one moves, in public, on a live dashboard.
Equity in a protocol is not a share certificate. It is the money that disappears first when something breaks.
Why did Moody’s put the capital ratio at 0.9%?
Because it measured loss-absorbing capital against exposure, the way it measures any financial institution.
Roughly $90M of tangible capital. Roughly $10B of managed exposure. That is 0.9%.
For reference, Basel III sets a 4.5% CET1 minimum, rising to 7% once the conservation buffer is counted.
Four things worth holding in your head at once.
- The ratio is not a claim about the USDS peg. Collateral still exceeds supply.
- Thin equity against a large book and insufficient backing are different questions.
- Moody’s named Sky Reserves as the central factor. The Foundation said publicly that it agrees.
- Both ratings carry a stable outlook. Neither is investment grade.
Sit with the strange part. A protocol published a number that makes it look thinly capitalised on a bank yardstick, then pointed everyone at the live dashboard where you can watch it move.
Most issuers would have buried it.
And this is not the first time a rating has rewritten the balance sheet. After S&P flagged the small surplus buffer last year, Sky Governance voted on 14 March 2026 to cut token buybacks from 75% of surplus to 7.5% and push almost everything else into reserves.
Get Mihawk’s stories in your inbox
Join Medium for free to get updates from this writer.
The analyst wrote a note. Governance rewrote the waterfall. That feedback loop barely existed in DeFi two years ago.
How do junior and senior risk capital work onchain?
This is the piece credit people recognise instantly and most crypto readers have never seen.
Allocators in the Sky Agent Network do not simply borrow and disappear. They post first-loss capital before they deploy anything.
- Junior Risk Capital is the allocator’s own treasury, or capital rented from outside investors. It burns first.
- Senior Risk Capital is the USDS drawn from the protocol. It is only exposed once the junior layer is gone.
- How much junior capital is required comes from a Capital Requirement Ratio that varies by asset class.
- Those ratios adapt their logic from the Basel Accords, scored across financial risk, smart contract risk and administrative risk.
- Fall below the requirement and the monthly settlement cycle applies penalties.
That is securitisation tranching, running on a public ledger, with the first-loss position disclosed rather than buried in an appendix.
For context on how rare this still is: onchain tranching accounts for less than 1% of DeFi lending deposits today.
Where does protocol surplus go before anyone gets paid?
There is a published answer, ratified on 13 August 2026. Every settlement cycle splits surplus like this.
- 50% to the Surplus Buffer
- 22.5% to SKY buybacks funding staking rewards
- 22.5% to USDS staking rewards
- 5% to SKY buy and burn
Read the order rather than the percentages. Loss absorption is funded before distributions. Sky Governance can change that split, and has already done so twice this year.
For scale: Sky Protocol generated $107.35M in Gross Protocol Revenue in Q2 2026, and $231.66M across the first half.
Where do sUSDS and the Sky Savings Rate sit in the stack?
Near the top. That is the whole proposition.
sUSDS is how you hold the Sky Savings Rate. Supply USDS, receive sUSDS, and the exchange rate compounds in your favour.
What makes that claim senior:
- The rate is funded from aggregate protocol revenue, not from one strategy or one borrower.
- You are not exposed to any single agent, pool or collateral type.
- It is governance-set and variable, published live rather than promised in advance.
- No lockups, no exit fees. Convert back to USDS whenever you want.
Then look at the three rates the protocol prints at once.
- sUSDS at 3.52%
- stUSDS at 5.07%, which lends against SKY collateral
- Staked SKY at 6.57%
That is not a product menu. That is a yield curve for seniority. The safest claim pays the least, exactly as it should.
And the stack is still being built. On 8 October 2026, governance launched the stUSDS Keeper, formalising a rung that did not exist a year ago. New instruments, added by vote, mid-quarter.
Is institutional capital actually pricing any of this?
Follow the balance sheets instead of the press releases.
- Galaxy added $100M of sUSDS to its corporate treasury on 23 September 2026 and approved it as collateral across its institutional trading business.
- sUSDS closed Q2 2026 at $5.52B, up 149% year over year.
- Sky Protocol is, so far, the only stablecoin protocol formally rated by both Moody’s and S&P.
Nobody puts $100M into a yield-bearing stablecoin because the APY looks nice.
They do it because they can finally locate themselves in the capital structure and know who absorbs the loss ahead of them.
What should you actually watch from here?
Four numbers. All public, all updated without waiting for a quarter to close.
- Protocol Collateral against stablecoin supply. That gap is the cushion.
- Sky Reserves against the $150M target. Loss absorption before distributions.
- Gross Protocol Revenue, quarter over quarter.
- Whether the surplus split changes, and who voted for it.
Equity and debt were invented for entities that can be sued, wound up and liquidated. A protocol cannot be any of those things.
So the industry is rebuilding the stack from first principles, in public, with a rating agency now reading over its shoulder.
Here is where I would like your take.
If a protocol’s collateral comfortably exceeds its liabilities, does a 0.9% capital ratio actually matter? Or is that a bank metric being forced onto something that was never built to be a bank?
Tell me below. I read every response, and I will argue with the good ones.
