“Control beats filing” is usually stated as a priority rule: a control-perfected interest ranks ahead of a filing-perfected one. That’s true, and it’s the headline. But it undersells what’s actually going on, because it makes the two methods sound like different rungs on the same ladder — as if filing is simply a weaker version of control.
They are not the same kind of thing at all. Filing and control protect a lender in fundamentally different ways, and for an asset like Bitcoin, only one of them actually does the job. Understanding why control beats filing — not just that it does — is what separates a lender who’s checking a box from one who understands what their security interest is made of.
What filing actually does, and doesn’t do
Perfection by filing is perfection by notice. You register a financing statement in a public registry, and that registration tells the world: this lender has a claim on this collateral. Anyone searching the registry can find it. Priority, traditionally, runs by the clock — first to file, first in line.
Notice is genuinely useful. It resolves disputes between competing secured parties by establishing who staked their claim first. But notice has a structural limit that matters enormously for an asset that can move in seconds: filing tells the world about your claim, but it does nothing to stop the collateral from moving.
This is the point that the priority framing obscures. A lender perfected only by filing has a registered claim — and a borrower who can still transfer the Bitcoin to a third party. The registration doesn’t sit between the borrower and the asset. It’s a notice on a public record, not a hand on the collateral. For traditional assets this gap is tolerable, because the assets are slow, illiquid, or hard to move without the transfer itself creating a record. For Bitcoin, the gap is the whole problem: the asset is bearer, it settles globally in minutes, and a borrower with the keys can move it before anyone refreshes a registry search. Filing-only perfection against Bitcoin is a notice about an asset that has already left.
Under Canadian law, this is exactly where directly-held Bitcoin lands by default. Absent a specific statutory category, it falls into the catch-all class of “intangibles” — the least negotiable form of collateral — which is perfected by registration, not control. So the default legal treatment of Bitcoin as collateral hands the lender the weaker of the two protections, the one that gives notice but no actual power over the asset.
What control does that filing can’t
Perfection by control is perfection by power. Instead of registering a claim about the asset, the lender obtains the practical ability to govern the asset itself — to participate in any transfer, and to prevent the borrower from moving it unilaterally.
This is a categorically stronger position, and the strength is operational before it’s legal. A lender with control isn’t relying on a record that warns other parties; the lender is structurally positioned so the collateral cannot leave without its participation. The borrower can’t transfer the Bitcoin to a third party, because the transfer requires authorization the borrower doesn’t solely hold. Where filing said “everyone is on notice that I have a claim,” control says “the asset cannot move without me.”
For Bitcoin specifically, this is the only protection that fits the asset. The thing that makes filing inadequate — that the asset is fast, bearer, and easily transferred — is precisely the thing control addresses. Control meets the asset on its own terms: it answers cryptographic power with legal-and-operational power over the same keys.
And this is why the priority rule comes out the way it does. A control-perfected interest doesn’t rank ahead of a filing-perfected one because of some arbitrary statutory preference. It ranks ahead because control is the stronger relationship to the asset. The lender who can prevent the collateral from moving has a better claim than the lender who merely told the registry about it — even if the registry filing came first. The priority rule is just the law recognizing the difference in substance.
How a lender actually establishes control
The mechanism that delivers control in practice is the multi-signature arrangement, and it’s worth being concrete about how it works, because this is where the legal concept and the operational structure become the same thing.
In a two-of-three multi-signature setup, three keys exist — typically one held by the lender, one by a qualified custodian, one by the borrower — and any transaction requires two of the three signatures. The borrower cannot move the collateral alone, because the borrower holds only one key. The lender’s participation is required for any disposition the borrower might attempt. That is control in the operational sense: the lender sits in the transfer path and can’t be routed around.
The legal frameworks that recognize control map onto exactly this structure. Control is satisfied where the secured party’s participation is required for any disposition and the debtor cannot transfer unilaterally — which is precisely the two-of-three arrangement. The legal test and the custody design describe the same thing. This is the convergence worth internalizing: the multi-signature structure a careful lender would build for purely operational reasons — to make sure the borrower can’t run off with the collateral — is the same structure that perfects the security interest by control and earns it priority. You don’t build custody one way and perfection another. One well-designed structure does both.
There’s a deeper logic here too. The legal definition of control fails in exactly the case where the borrower retains unilateral power over the collateral — and that’s the same case sound lending already refuses to allow. The law’s requirement and the lender’s discipline encode the identical rule: never let the borrower be able to move the collateral alone. Good secured lending and valid control-perfection are not two constraints. They’re one constraint, stated in two vocabularies.
What this means in practice
For a lender, the instruction that falls out of all this is straightforward, and it should be a condition of origination rather than an optional enhancement: structure the custody so you have control, not just a filing.
Where the law has adopted a control framework, this gives you priority and the practical power to stop the collateral from moving — both, from one structure. Where the law hasn’t fully caught up, building for control rather than relying on registration alone is still the stronger position, because the operational reality — the borrower can’t move the asset without you — holds regardless of how settled the statute is. Control is better collateral protection on the merits, before it’s ever better on priority.
The reason “control beats filing” matters is therefore not that the law happens to rank one above the other. It’s that for a bearer asset that settles in minutes, a notice in a registry was never going to be enough. The only protection that actually secures the collateral is the one that puts the lender in the transfer path. Filing tells the world you have a claim. Control means the collateral can’t leave without you. Against Bitcoin, only the second one is really collateral.