The number is 2%. Not 8%. Not 12%. Two.
A Nasdaq-listed entity called PowerCompute has refinanced $18 million in debt using Bitcoin as collateral at an initial interest rate of roughly 2%. In a market where Bitcoin-backed loans have historically priced between 8% and 15% annualized, that single data point is doing more narrative heavy-lifting than the entire transaction warrants.
I have spent the past decade auditing lending protocols and modeling collateralized debt structures. One rule has never failed me: a rate that far below market is either a subsidy, a trap, or evidence that the lender believes the collateral is better than the market does. The first case burns the lender. The second burns the borrower. The third is the only scenario where this deal is what it appears to be.
So which one is this?
Context: Bitcoin Collateral Is Not Innovation
Let me clear the technical underbrush first. Bitcoin-backed lending is a mature product. It has existed since 2018, when Genesis and BlockFi began processing collateralized loans for miners and institutional holders. Ledn, Unchained Capital, and Galaxy Digital all offer variations of the same structure: pledge BTC, borrow dollars, maintain a loan-to-value ratio that keeps the lender solvent through a drawdown.
Nothing about this category is frontier technology. Bitcoin has no native smart contract capability, so the collateral must be held by a third party — an institutional custodian, an MPC wallet arrangement, or a Discreet Log Contract structure that pre-signs transactions to enforce repayment. The technical differentiation between these options is enormous. A DLC structure reduces counterparty risk but requires pre-committed transaction paths. A centralized custodian is simpler but concentrates risk in the exact institutions that failed during the 2022 credit crisis.
The source material does not disclose which custody model this loan uses. That omission is not a detail. It is the story.
The other distinguishing feature here is the borrower. PowerCompute's Nasdaq listing introduces disclosure obligations, auditor scrutiny, and a public equity market that will react violently when Bitcoin moves 30% in a week. That is the genuinely new element: not the loan, but the entity taking it.
Core: The Three Constraints Nobody Verified
Let me walk through the actual risk stack, constraint by constraint.
Constraint One: Custody.
Somewhere — and we do not know where — PowerCompute's Bitcoin sits. The options range from a regulated trust company like BitGo or Coinbase Custody to an offshore entity with ambiguous legal standing in a U.S. bankruptcy proceeding.
The 2022 cycle provided the precedent. BlockFi and Celsius held billions in user assets through custody arrangements that looked institutional on the marketing deck and turned out to be intercompany loans in the liquidation report. Trust is not a feature. Verification is.
t trust, verify the stack. The stack here is an unverified black box.
Constraint Two: The Liquidation Math.
Here is where the numbers get uncomfortable. Bitcoin-backed loans are priced around a loan-to-value ratio — the percentage of the collateral's market value that the borrower can draw. If the LTV is 50%, PowerCompute's $18 million loan requires roughly $36 million in pledged Bitcoin. If the LTV is 70% — aggressive, but not unheard of in this market — the collateral requirement drops to about $25.7 million.
Now apply the historical volatility. I track Bitcoin's drawdown distribution closely because it determines liquidation probability. In 2022, Bitcoin fell from $48,000 to $15,500 — a 68% decline. A borrower with a 50% LTV and no ability to post additional margin would have been liquidated during that move. The entire collateral position would have been sold at the worst possible price, converting a temporary drawdown into a permanent loss.
The 2% interest rate suggests the lender demanded substantial overcollateralization. You do not lend at 2% to a borrower with a 70% LTV. But "suggests" is not "verifies." The LTV, the liquidation threshold, the margin call mechanism — none of it has been disclosed. This is not diligence. This is a photograph of a balance sheet with the numbers blurred out.
Math has no mercy. If Bitcoin corrects 40% from here and the loan carries a 50% LTV with a 60% liquidation threshold, PowerCompute faces a margin call at roughly a 16.7% decline. That is not a tail risk scenario. That is a routine Bitcoin Tuesday.
Constraint Three: The Rate That Shouldn't Exist.
The word "initial" is doing invisible work in this headline. The rate is not 2%. The rate is 2% initially. There is a re-pricing event somewhere in this loan's future, and its terms determine whether this is a smart refinancing or a two-year countdown to a funding shock.
The economics are straightforward. If the lender sourced capital at a market rate of 5-7% and lent it at 2%, the transaction loses money before any operating costs. Institutions do not do that without a strategic reason: customer acquisition, cross-selling, or a promotional rate designed to establish a market benchmark. All three imply that the 2% rate is temporary.
What happens at re-pricing? If the loan converts to a floating rate tied to a benchmark, PowerCompute's interest expense could double or triple. The $18 million position is small enough that the absolute dollar impact is modest — but as a signal, the "initial" qualifier voids the most important marketing point of the entire announcement.
The Borrower's Incentive Structure
Let me also examine the counter-party incentives, because this is where forensic analysis usually finds the body.
PowerCompute chose to pledge Bitcoin rather than sell it. That reveals a directional bet: management expects Bitcoin's upside to exceed the cost of the loan. If they expected flat or declining prices, the rational move would have been liquidation and cash retention. This is not a neutral treasury operation. It is a levered long position dressed in conservative clothing.
The refinancing also avoids equity dilution. Raising $18 million through a stock offering would have diluted existing shareholders. A loan, even a Bitcoin-backed one, keeps the capitalization table intact. As a capital structure decision, it makes sense. As a risk decision, it converts equity risk into liquidation risk — and liquidation risk is the one category that can wipe out the entire collateral position in a single oracle update.
During the Terra collapse in 2022, I watched the UST depeg cascade through the lending market in three days. The patterns were symmetrical: over-leveraged collateral, opaque redemption mechanics, and borrowers who discovered their "safe" loan terms contained structural triggers they never modeled. Based on my experience auditing these arrangements, I can tell you what matters: not the marketing rate, but the liquidation threshold, the re-pricing schedule, and the identity of the custody counterparty. None are public here.
There is also the regulatory layer. As a Nasdaq issuer, PowerCompute must disclose material financial arrangements in SEC filings. If this loan carries a variable rate tied to Bitcoin's price performance or a re-pricing event tied to the collateral's market value, the auditor may demand impairment testing on the pledged assets. The accounting treatment of the Bitcoin collateral — fair value versus cost method — will determine whether a 30% drawdown produces a quarterly earnings hit. That is not a theoretical concern. That is a 10-Q disclosure waiting to happen.
The market context matters too. We are in a sideways regime. Chop is for positioning, not for momentum. A single $18 million loan will not move Bitcoin's price, but it reshapes the positioning calculus for every corporate treasurer watching the rates desk. In a consolidation market, this kind of low-cost leverage signal carries more weight per unit of capital than it would during a parabolic rally — because institutions are hunting for carry, not beta.
Contrarian: What The Bulls Got Right
The skeptics — and I include myself in that camp — have a blind spot. The signal effect of this transaction is real, and its magnitude is larger than the $18 million suggests.
This is not just another miner taking a BTC loan. This is a Nasdaq-listed operating company using Bitcoin as balance sheet leverage. That is a different category of adoption. MicroStrategy showed that public companies can hold Bitcoin. PowerCompute's loan demonstrates a more advanced use case: Bitcoin as a working capital instrument that reduces the cost of debt.
If the 2% rate becomes a reference point — even a promotional one — it will distort the pricing benchmark for the entire Bitcoin lending market. Other listed companies with treasury positions will ask their CFOs why they are paying 12% on unsecured credit lines when a rival borrowed at 2% against BTC. Lenders will respond with new products. The market infrastructure sector — custody, MPC wallets, DLC implementations — benefits regardless of whether this specific loan performs well.
The capital flow direction also matters. PowerCompute pledged Bitcoin rather than selling it. Every Bitcoin that moves into collateralized storage is Bitcoin that is not hitting the sell side of the order book. In an environment where supply liquidity is already constrained, the cumulative effect of corporate collateralization is a structural bid.
Takeaway: Watch The Filings, Not The Headlines
This transaction is not a revolution. It is an $18 million refinancing with an eye-catching rate and no disclosed technical terms. The market will move on before the first margin call — or the first re-pricing announcement — forces the real story into the light.
I will be watching three signals. First, the SEC 8-K and 10-Q filings, which will reveal the loan's LTV, custody arrangement, and re-pricing schedule. Second, PowerCompute's behavior if Bitcoin corrects more than 25%: do they post additional collateral, or do they negotiate? Third, the broader market — whether other listed companies announce similar facilities in the next two quarters.
High yield, high graveyard. Low yield with undisclosed liquidation terms is just a slower burial. The 2% is not the story. The missing 98% of the contract is the story.
