Testnet Connect Wallet

What makes a tokenized security transfer compliant?

Where does the rule live? Who can change it? When does it run? What if it says no? It's totally compliant! Great. Which of the four? FOUR QUESTIONS, NOT ONE

A vendor conversation can answer one of these and let you assume the rest.

A tokenized security transfer is compliant when the rule about who may hold the asset is enforced at the moment ownership changes, by something the buyer cannot route around. Everything else is documentation of an intention. The difficult part was never writing the rule, it is deciding where the rule lives and who is allowed to change it.

If you are an issuer, a fund administrator, or the person who signs the transfer agent letter, this is written for you rather than for a developer. There is no code in it.

In plain terms. A compliant transfer of a tokenized security is one where each change of ownership is checked against the issuer's eligibility rules before it completes, the check cannot be skipped by dealing with the asset directly, and the outcome, allowed or refused and for which reason, is recorded somewhere an auditor can retrieve it later.

Why is "is it compliant?" the wrong question?

Because it has four answers, and a platform can give you a confident yes while meaning only one of them. In our experience of these conversations, the yes usually refers to the first question and quietly leaves the other three to your imagination.

Ask them separately and the picture changes. Here they are, with what each one actually decides.

The questionWhat it decidesWhat goes wrong when nobody asks it
Where does the rule live? Whether the rule can be skipped. A restriction enforced in a web application is optional for anyone willing not to use the web application.
Who can change it? Whether the rule is stable, and whose signature stands behind a change. An eligibility list that one administrator can edit without anyone noticing is a governance arrangement wearing a technical costume.
When does it run? Whether an ineligible transfer is prevented or merely discovered afterwards. Screening that happens before a trade catches the trades you were told about.
What happens when it says no? Whether you have an audit trail or a support ticket. A transfer that fails without naming a reason is indistinguishable from a bug, to your investor and to your regulator.

Where should the eligibility rule live?

There are three honest places to put it, and they are not equivalent.

In your books. The register remains the legal record, and for many issuers it always will be. This works, and it is what transfer agents have done for a century. What it does not do is stop anything in real time. It reconciles after the fact.

In your application. This is a common arrangement in tokenization products, and it is the one worth probing hardest. The question to put to any vendor is short: what happens if an investor interacts with the asset without opening your interface? If the answer involves the interface at all, the rule is a property of a web page.

In the asset itself. The rule travels with the token, and the check runs as part of the transfer rather than alongside it. This is the design intent behind the token standards written for regulated assets, of which ERC-3643 is the best known. It is also the model we build on, which we describe in more detail on our compliance at the transfer page.

None of this is exotic or new. It is the same question a custodian asks about segregation: not whether a policy exists, but whether the mechanics make it possible to breach.

Who is allowed to change the rule?

This is the question that gets skipped, and it is the one your counsel will care about most once it is pointed out.

Every workable system has privileged roles. Somebody has to add an investor once their onboarding completes, suspend transfers if an order requires it, and correct mistakes. The issue is never that these powers exist. It is whether you can produce a list of them.

Four things worth asking for in writing, before signing anything:

  • Which roles hold administrative power over transfer eligibility, named individually rather than as "the platform".
  • What each role can do without a second approval.
  • How a change is recorded, and whether it is observable by someone who is not the administrator.
  • Whether any role can exempt itself from the rules it administers.

That last one deserves a moment. A rule set where the administrator can quietly place themselves outside the rule is a different control from one where they cannot, even though both look identical in a product demonstration. It is a fair question to ask, and a good vendor will have a clear answer rather than a wounded one.

When does the check actually run?

Pre-trade screening is valuable and you should still do it. Its limitation is structural rather than technical: it can only examine transactions it is shown. Once an asset can move between two parties without an intermediary, the only moment guaranteed to occur on every transfer is the transfer itself.

This is why the question of timing collapses back into the question of place. A check that runs inside the transfer runs on all of them, including the ones nobody told you about, including the ones that happen at three in the morning through an interface you have never seen.

What happens when a transfer is refused?

A refusal is not a failure of the system, it is the system working. What matters is what the refusal leaves behind.

Three outcomes are common, and they are operationally very different. The transfer can fail silently, leaving your investor with a confusing error and your support desk with nothing to work from. It can fail with a generic message, which is marginally better. Or it can fail while naming the specific rule that refused it, which turns an angry phone call into a two minute explanation and gives you a record you can point at later.

For a regulated asset, ask to see what a refusal looks like. Not a description of it. The actual output.

Does the rule survive the asset moving between chains?

If your asset will exist on more than one network, and increasingly it will because your investors are not all in the same place, this becomes a fifth question hiding inside the first one.

An asset issued on one network and represented on another is, mechanically, two deployments. Each enforces whatever rules are attached to it. Compliance on the first does not travel to the second by itself, and a wrapped representation with no rules attached is an ordinary token wearing your asset's name.

The thing to ask for is not a promise that it works. It is a description of what enforces eligibility on the destination network, in that network's own terms. We use Chainlink for the cross-chain messaging and the Automated Compliance Engine for the policy layer, and our cross-chain infrastructure for tokenized assets page sets out how the pieces fit. Whatever a platform uses, the answer should be specific enough to check.

A checklist for your next vendor call

Copy this. It takes about fifteen minutes to work through and it separates products quickly.

  • What happens if an investor interacts with the asset without using your interface?
  • Show me a transfer being refused, and tell me what the refusal says.
  • List every privileged role and what each can do unilaterally.
  • Can any administrator exempt themselves from the transfer rules?
  • How is a change to the eligibility rules recorded, and who can see it?
  • If the asset exists on a second network, what enforces eligibility there?
  • Can I verify any of this myself, without asking you?
  • What does the check cost per transfer, and who pays it?

The last two matter more than they look. A platform that can point you at something you can inspect independently is making a different kind of claim from one that cannot. And enforcement that runs on every transfer is not free, so a vendor who says there is no cost has either solved something remarkable or has not measured it.

The one thing to take away

Compliance is not a feature that a product either has or lacks. It is a set of answers about mechanism, and the useful ones are specific: this rule, enforced in this place, changeable by these people, producing this record when it refuses. If the answers you get are about intentions and roadmaps rather than mechanism, you have learned something valuable anyway.

Common questions

Is a security token the same thing as a tokenized security? The terms are used interchangeably in practice, and both describe an instrument whose ownership is recorded on a blockchain while remaining subject to the securities rules of the relevant jurisdictions. The underlying obligations do not change because the register does. Background reading on the broader category sits under security token offering.

Do I need ERC-3643 specifically? Not necessarily. What you need is transfer eligibility enforced by the asset rather than by an application. ERC-3643 is one well documented way to achieve that, and there are others, including policy engines attached to a token that let the rule set change without reissuing the asset. The standard matters less than where the enforcement sits.

Does checking every transfer make the asset expensive to move? It adds a real cost to each transfer. For a regulated asset that cost is usually a sensible trade, because the alternative is a restriction that can be stepped around. For an unregulated asset it often is not.

Can an issuer still freeze or recover tokens? Controls of that kind are part of designs built for regulated assets. In ours they are exercised through the same layer that handles eligibility. They raise their own governance questions, in particular the "who can change it" question above, and they deserve their own discussion with your counsel.

Can I check a platform's claims without taking their word for it? Often yes. Contracts deployed to public networks can be source-verified and read by anyone, and a platform confident in its design will point you at them. Ours are indexed in a public security repository, and the testnet application is open to anyone: after a quick verification step, you can try a transfer and watch what happens.

Testnet, unaudited. Not a solicitation. Not an offer of securities. Nothing here is legal advice. Every transaction you make on this site is signed by your own wallet on a public testnet. No real money moves.