The trading desk will take Bitcoin as collateral, the risk desk decides how to hold it
September 2026
Thomas Wu, chief financial officer at Ledn, examines how banks are approaching Bitcoin as collateral and why risk considerations, from capital treatment and custody to liquidity and liquidation, are shaping how institutions choose to hold the asset
Image: Ledn
Ask a bank’s financing desk whether it would lend against Bitcoin and, increasingly, the answer is yes. Ask the credit and risk desk whether Bitcoin clears the collateral bar, and the conversation gets longer. That split, the financing side ready and the risk side as the gate, is the real story of digital assets entering secured funding. The question is no longer whether an institution will take Bitcoin as collateral. It is who inside the institution decides what to accept, and on what terms.
When institutions decide to underwrite an asset class, the tell is not the headlines about banks warming to crypto. It is the structures they are choosing to hold it through. When you look at how the largest institutions are actually accepting Bitcoin today, almost none of them are holding the asset directly. They are routing around it, and the route they pick tells you which desk is still apprehensive.
Two banks, two workarounds
Take the two most visible moves. In June, Morgan Stanley began letting wealth clients convert their Bitcoin in-kind into shares of a spot Bitcoin exchange traded product (ETP), then borrow against those shares through a referral arrangement with a digital asset firm. Read that carefully: the collateral the bank is ultimately lending against is not Bitcoin. It is a security, the ETP wrapper, that happens to track Bitcoin. The native asset has to be turned into a familiar container before the balance sheet will touch it.
J.P. Morgan has taken a different route to the same destination. Late last year, its reported programme aimed at institutional clients was to accept the native token as collateral, but it is built so the bank never owns the underlying asset, not even on default.
The pledged asset sits with a third-party custodian, and if a borrower defaults, third-party partners handle the liquidations. In ordinary asset-based lending, a bank will temporarily hold the defaulted asset to sell, whereas this programme is specifically engineered to never hold Bitcoin directly, because under the banking capital rules, owning digital assets is an expensive event. The bank gets exposure to the collateral without ever holding the asset itself, managing counterparty and operational risk by keeping it at arm’s length.
Two of the most sophisticated institutions in the world, two different answers to the same eligibility question, and both engineered to avoid holding Bitcoin directly. That is not indecision, and it is not simply internal nerves. It is two firms solving for the same underlying constraint, which is largely a regulatory one: the punitive capital treatment that still attaches to holding Bitcoin outright, on top of balance sheet and custody considerations. The risk desk is where that rulebook meets the asset, and its job is to find a structure the rules will bless. The wrapper and the third-party custodian are not conveniences. They are the price of getting the collateral past it.
What the risk officer is actually testing for
It is worth being precise about what that desk is assessing, because it is not the thing most crypto commentary assumes. A credit officer is not debating whether Bitcoin will go up or down. They are testing whether the collateral can be valued continuously, controlled legally, and liquidated under stress without the recovery evaporating. Those are the questions a collateral schedule is built to answer, and the frameworks that answer them were written for equities, receivables, and real estate. Applied to Bitcoin, those frameworks can misread what is in front of them. The instinct to wrap Bitcoin in a security, or perfect it at a distance across jurisdictions, is partly a way of making an unfamiliar asset legible to a familiar model, even when that model is miscalibrated on the liquidity dimension, where Bitcoin is genuinely strong.
Here is the point that tends to surprise a finance audience. On the measures a collateral desk actually cares about — continuous market depth, tight spreads, and the ability to liquidate size under stress — Bitcoin now ranks among the most liquid assets available to pledge.
On any 24-hour view, it sits behind foreign exchange, US Treasuries, and the largest index ETFs, and on par with individual mega-cap equities that many of the same desks accept without a second thought.
The global asset trades every hour of every day, which means the global asset trades at all hours somewhere in the world, so any overnight gap in which a position cannot be defended is minimal.
This is not a promotional claim. It is a collateral-quality argument, and it is one that institutional credit investors have made themselves when they have looked closely.
The relevant question for a risk desk is not Bitcoin’s price volatility in isolation, since every collateral type has volatility, which is what haircuts exist to absorb. It is whether, when a liquidation is triggered, the asset can be sold quickly enough and deep enough to make the lender whole.
Bitcoin has now been tested on exactly that, through market liquidation cascades that would have gapped a thinner asset: roughly US$19 billion of forced liquidations on 10 October 2025, and some US$2.5 billion in February 2026. In both, the market absorbed the selling pressure and cleared, and secured lenders margining against Bitcoin came through with principals intact.
What changes if you hold it directly
There is a third route, and it is the one that most closely resembles how the finance world already manages collateral: hold the native asset and margin it in-house continuously.
Firms that operate this way monitor loan-to-value in real time, issue margin calls automatically as thresholds approach, and liquidate against the collateral directly rather than through a wrapper.
That model carries its own demands. It requires collateral management capability, real-time risk systems, and a liquidation record that can be shown rather than asserted.
That record is now being tested by parties whose job is scepticism. When a book of Bitcoin-backed consumer loans run on this model was brought to the asset-backed securities (ABS) market this year, it carried an investment-grade rating from S&P — a first for a digital asset company.
The instructive part is not the rating itself but what the analysis rewarded. In a conventional securitisation, forced asset sales are a distress signal, the thing the structure is built to avoid. Here, rule-based margining and automated liquidation were the credit protection: the mechanism that turns collateral into cash and caps losses before they compound, rather than leaving the decision to discretion at the worst possible moment. The behaviour that reads as alarming in a headline — price swings triggering liquidations — is precisely what the noteholder is relying on. Borrowers top up, or they repay, and the exposure de-risks itself.
But it removes the layers that the wrapper adds, and with them the cost, the basis risk between the asset and its container, and the operational dependency on a chain of intermediaries. As more institutions cross the line from evaluating Bitcoin collateral to holding it, the question they will face is the one banks have always faced with any new asset class: not whether to take it, but how close to it they are willing to sit.
Where this goes
The trajectory is familiar to anyone who has watched a collateral type mature. The first movers accept it through the most conservative structure available, because that is what gets it past risk. As the empirical liquidation evidence accumulates and the custody question is answered, the structures simplify, and the desks move closer to the asset. Bitcoin is early in that arc for banks, held today mostly through wrappers and intermediaries. But the direction of travel, from routing around the asset to holding and margining it directly, is the one every previous form of collateral has taken.
The trading desk decided it would take Bitcoin some time ago. The interesting work now sits on the risk desk, deciding not whether it counts, but how to hold it.
When institutions decide to underwrite an asset class, the tell is not the headlines about banks warming to crypto. It is the structures they are choosing to hold it through. When you look at how the largest institutions are actually accepting Bitcoin today, almost none of them are holding the asset directly. They are routing around it, and the route they pick tells you which desk is still apprehensive.
Two banks, two workarounds
Take the two most visible moves. In June, Morgan Stanley began letting wealth clients convert their Bitcoin in-kind into shares of a spot Bitcoin exchange traded product (ETP), then borrow against those shares through a referral arrangement with a digital asset firm. Read that carefully: the collateral the bank is ultimately lending against is not Bitcoin. It is a security, the ETP wrapper, that happens to track Bitcoin. The native asset has to be turned into a familiar container before the balance sheet will touch it.
J.P. Morgan has taken a different route to the same destination. Late last year, its reported programme aimed at institutional clients was to accept the native token as collateral, but it is built so the bank never owns the underlying asset, not even on default.
The pledged asset sits with a third-party custodian, and if a borrower defaults, third-party partners handle the liquidations. In ordinary asset-based lending, a bank will temporarily hold the defaulted asset to sell, whereas this programme is specifically engineered to never hold Bitcoin directly, because under the banking capital rules, owning digital assets is an expensive event. The bank gets exposure to the collateral without ever holding the asset itself, managing counterparty and operational risk by keeping it at arm’s length.
Two of the most sophisticated institutions in the world, two different answers to the same eligibility question, and both engineered to avoid holding Bitcoin directly. That is not indecision, and it is not simply internal nerves. It is two firms solving for the same underlying constraint, which is largely a regulatory one: the punitive capital treatment that still attaches to holding Bitcoin outright, on top of balance sheet and custody considerations. The risk desk is where that rulebook meets the asset, and its job is to find a structure the rules will bless. The wrapper and the third-party custodian are not conveniences. They are the price of getting the collateral past it.
What the risk officer is actually testing for
It is worth being precise about what that desk is assessing, because it is not the thing most crypto commentary assumes. A credit officer is not debating whether Bitcoin will go up or down. They are testing whether the collateral can be valued continuously, controlled legally, and liquidated under stress without the recovery evaporating. Those are the questions a collateral schedule is built to answer, and the frameworks that answer them were written for equities, receivables, and real estate. Applied to Bitcoin, those frameworks can misread what is in front of them. The instinct to wrap Bitcoin in a security, or perfect it at a distance across jurisdictions, is partly a way of making an unfamiliar asset legible to a familiar model, even when that model is miscalibrated on the liquidity dimension, where Bitcoin is genuinely strong.
Here is the point that tends to surprise a finance audience. On the measures a collateral desk actually cares about — continuous market depth, tight spreads, and the ability to liquidate size under stress — Bitcoin now ranks among the most liquid assets available to pledge.
On any 24-hour view, it sits behind foreign exchange, US Treasuries, and the largest index ETFs, and on par with individual mega-cap equities that many of the same desks accept without a second thought.
The global asset trades every hour of every day, which means the global asset trades at all hours somewhere in the world, so any overnight gap in which a position cannot be defended is minimal.
This is not a promotional claim. It is a collateral-quality argument, and it is one that institutional credit investors have made themselves when they have looked closely.
The relevant question for a risk desk is not Bitcoin’s price volatility in isolation, since every collateral type has volatility, which is what haircuts exist to absorb. It is whether, when a liquidation is triggered, the asset can be sold quickly enough and deep enough to make the lender whole.
Bitcoin has now been tested on exactly that, through market liquidation cascades that would have gapped a thinner asset: roughly US$19 billion of forced liquidations on 10 October 2025, and some US$2.5 billion in February 2026. In both, the market absorbed the selling pressure and cleared, and secured lenders margining against Bitcoin came through with principals intact.
What changes if you hold it directly
There is a third route, and it is the one that most closely resembles how the finance world already manages collateral: hold the native asset and margin it in-house continuously.
Firms that operate this way monitor loan-to-value in real time, issue margin calls automatically as thresholds approach, and liquidate against the collateral directly rather than through a wrapper.
That model carries its own demands. It requires collateral management capability, real-time risk systems, and a liquidation record that can be shown rather than asserted.
That record is now being tested by parties whose job is scepticism. When a book of Bitcoin-backed consumer loans run on this model was brought to the asset-backed securities (ABS) market this year, it carried an investment-grade rating from S&P — a first for a digital asset company.
The instructive part is not the rating itself but what the analysis rewarded. In a conventional securitisation, forced asset sales are a distress signal, the thing the structure is built to avoid. Here, rule-based margining and automated liquidation were the credit protection: the mechanism that turns collateral into cash and caps losses before they compound, rather than leaving the decision to discretion at the worst possible moment. The behaviour that reads as alarming in a headline — price swings triggering liquidations — is precisely what the noteholder is relying on. Borrowers top up, or they repay, and the exposure de-risks itself.
But it removes the layers that the wrapper adds, and with them the cost, the basis risk between the asset and its container, and the operational dependency on a chain of intermediaries. As more institutions cross the line from evaluating Bitcoin collateral to holding it, the question they will face is the one banks have always faced with any new asset class: not whether to take it, but how close to it they are willing to sit.
Where this goes
The trajectory is familiar to anyone who has watched a collateral type mature. The first movers accept it through the most conservative structure available, because that is what gets it past risk. As the empirical liquidation evidence accumulates and the custody question is answered, the structures simplify, and the desks move closer to the asset. Bitcoin is early in that arc for banks, held today mostly through wrappers and intermediaries. But the direction of travel, from routing around the asset to holding and margining it directly, is the one every previous form of collateral has taken.
The trading desk decided it would take Bitcoin some time ago. The interesting work now sits on the risk desk, deciding not whether it counts, but how to hold it.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one digital assets news source this year, make sure it is your free subscription to The Digital Assets Edge
100% ON RETURNS If you invest in only one digital assets news source this year, make sure it is your free subscription to The Digital Assets Edge
