Non-equivalent, not absent — Russia's digital-asset regime in comparative frame

Executive summary

There are two lazy readings of Russia's digital-asset position, and both cause control failures. The first is that the activity is unregulated. It is not: Federal Law No. 259-FZ created a statutory category for digital financial assets in 2020, the Bank of Russia maintains registers of the platform operators inside that perimeter, and a subsequent experimental legal regime extended defined activity under supervision. The second reading is that because a supervised regime exists, an institution can rely on it the way it might rely on an EU or Singaporean authorisation. It cannot.

The useful frame is neither presence nor absence but equivalence. Measured against MiCA Titles III and IV, the MAS single-currency stablecoin standard, the VARA fiat-referenced rulebook and US charter conditions, the Russian framework differs on three specific features that matter to a control estate: there is no single published reserve-composition standard applied uniformly across issuances, there is no statutory redemption floor operating independently of contract terms, and there is no mandated public attestation cadence. Those are not rhetorical points. They are the three places an institution's evidence file would otherwise be populated by someone else's rulebook.

The governance conclusion generalises well beyond one jurisdiction. When a home regime is non-equivalent rather than non-existent, the missing evidence does not disappear — it migrates onto the institution's own balance of controls. The practical response is a rail and counterparty eligibility register: assessed regimes, determinations, supporting evidence, assessment dates, and the exposures currently booked against each category. Institutions that hold that register can answer a supervisor in one meeting. Institutions that hold a policy statement instead will spend a quarter reconstructing one.

Why comparative, and why now

Two developments have made this a live diligence question rather than a desk-research one. Rouble-referenced token issuance has moved from concept to circulating instrument, and the experimental legal regime has been used to advance cross-border settlement use cases. Whenever a new fiat-referenced instrument begins circulating on a platform whose rulebook is not familiar to Western compliance functions, three questions arrive at the same time: can we touch it, can our counterparties touch it, and can we prove which answer we gave.

We are deliberately answering those in comparative form. A jurisdiction-specific risk narrative ages badly and invites imprecision. A comparative matrix against the four most detailed published regimes ages well, because the control questions it asks — who authorises, what backs it, what redemption right exists, what is disclosed and how often, who examines, and how the perimeter treats cross-border activity — are the same six questions for every regime an institution will ever assess. Russia is the current test case. The method is the deliverable.

The Russian architecture — DFA law and the experimental legal regime

Federal Law No. 259-FZ of 31 July 2020 established digital financial assets as rights recorded in an information system, covering monetary-claim and certain security-like rights, and made issuance and trading conditional on the platform being operated by an entity entered in a Bank of Russia register. The structural choice is important: the supervised object is primarily the platform operator, and the instrument's terms are set in the issuance decision and the platform rules rather than by a uniform product rulebook.

That is a registry-and-operator model. It is coherent, and it gives the supervisor a clear point of leverage. But it produces instrument-level variability that a product-rulebook model such as MiCA Title IV or the MAS stablecoin standard specifically removes. Under a product rulebook, an institution can make a class determination and then test issuer-level compliance. Under an operator-and-issuance-decision model, class determinations are unsafe, because two instruments on the same registered platform can carry materially different backing and redemption terms.

The experimental legal regime layered on top narrows participation — in significant part to qualified investors — while permitting defined activity, including settlement use cases, to be tested under supervision. Institutions frequently misread investor gating as a signal of prudential quality. It is not. Gating reduces conduct and suitability exposure. It says nothing about reserve composition, custody arrangement, settlement finality or operational resilience, which is where an institution's own loss would come from.

Rouble-referenced tokens and the reserve-attestation question

For any fiat-referenced token, in any jurisdiction, three questions determine whether the instrument is evidenced or merely asserted. What exactly backs it, held where and by whom. What right does a holder have to convert back to the referenced currency, on what timetable, and does that right survive stress. And who verifies the first two answers, how often, and is that verification public.

MiCA answers all three prescriptively for e-money tokens: prescribed high-quality liquid reserve composition with concentration and custody rules, redemption at par at any time without a defeating fee, and ongoing reporting with enhanced duties above significance thresholds. Singapore's framework answers them with a monthly independent attestation, an annual audit and a maximum redemption period. VARA answers them through rulebook-set reserve verification and periodic reporting. US practice has converged on monthly composition reporting through charter conditions and supervisory expectation rather than a single statute.

For rouble-referenced issuance under the Russian framework, the answers currently sit in the issuance decision and platform rules rather than in a uniform published standard with a mandated public attestation cadence. The consequence for a foreign institution is specific and not rhetorical: reserve adequacy cannot be inherited from the regime and must be established instrument by instrument, or the exposure declined. An institution that books such an exposure on the strength of the platform's registration has substituted a perimeter fact for a control.

The comparative matrix — five regimes, six control questions

The table below is deliberately narrow. It compares published frameworks on the six questions that determine what evidence an institution must generate itself. It is not a ranking, and it is not a recommendation about any instrument or counterparty.

Read down the reserves and redemption columns and the pattern is clear. The regimes that prescribe composition and a redemption floor transfer evidential weight from the institution to the rulebook. The regimes that leave those terms to contract leave the weight where it started. Neither arrangement is inherently unsound. They simply produce different diligence obligations, and an institution that treats them identically will under-document one of them.

Equivalence, substantial similarity, and non-equivalence

These three terms are used interchangeably in practice and should not be. Equivalence is a determination made by a regulator about another jurisdiction's regime, and it permits reliance. Substantial similarity, as we use it in the conduct spine, is an institution's own evidenced attestation that a control designed against one named regime satisfies the operative requirement of another — a method for building once and evidencing to several supervisors. Non-equivalence is simply the absence of the first, and it says nothing about quality.

The failure mode we see most often is a substantial-similarity attestation being written as though it were an equivalence determination. An institution maps its controls to a second regime, concludes the outcomes are comparable, and then files the conclusion as though a regulator had blessed the underlying regime. Those are different artefacts with different audiences. The attestation is about the institution's controls. The determination is about a jurisdiction. Only one of them is yours to make.

Writing a defensible determination

A defensible non-equivalence determination records four things: the specific features of the home regime assessed, the features found not comparable and against which reference regime, the compensating evidence the institution will generate itself, and the exposure ceiling that applies until that evidence exists. It is a page. Its absence is what turns a routine supervisory question into a finding.

Rail eligibility as a control, not a policy statement

Most institutions have a position on which settlement rails they will use. Fewer have a control. The difference is enforcement: a position lives in a policy document, while a control appears in onboarding logic, payment screening, custody permissions and distribution approvals, and produces a testable record when it blocks something.

Rail eligibility should be recorded per rail and per exposure type, with the finality basis named. Where finality rests on platform rules rather than statute or a recognised system designation, the institution's reconciliation cadence and dispute-resolution path carry the weight finality would otherwise carry, and both should be evidenced. Where a rail is ineligible, the register should show which control enforces that and when it was last tested. An ineligibility that has never fired and has never been tested is an assertion.

Counterparty and correspondent diligence under a non-equivalent regime

Direct counterparty registers are the easy part. The exposure that surprises institutions arrives through fourth-party dependencies: a platform operator, a custodian, an oracle, a data provider or a technology vendor sitting inside the transaction or evidence path of a counterparty the institution has already approved. Jurisdictional exposure enters through those dependencies far more often than through a direct relationship, and a counterparty-level register cannot see it.

The remedy is mapping rather than questionnaires. Where a dependency is material to settlement, custody or evidence production, it should appear in the third-party register with its own jurisdictional attribute, concentration position and substitution analysis. Institutions that have done this work discover concentrations they did not know they held, which is uncomfortable and useful in that order.

Sanctions obligations as a control category

We are treating this as a control category rather than a narrative, and the boundary is deliberate. The institution's obligations are to screen against the applicable consolidated lists at the frequency its risk assessment requires, to maintain originator and beneficiary information consistent with the FATF travel-rule standard where it applies, to escalate and file where thresholds are met, to evidence that screening covers every path into the institution rather than only account onboarding, and to test that coverage.

What we will not do on a public surface is characterise which parties are engaged in what conduct, or describe evasion methods. Designation and licensing are legal facts that change, and they are to be checked at source against the OFAC, EU Council and UK OFSI publications with counsel involved. An institution that takes its designation reading from commentary rather than from the consolidated lists has created a control weakness in the act of trying to close one.

The control question that generalises is coverage, not geography. Screening that runs on onboarding but not on payment instruction, or on payment instruction but not on custody transfer or token distribution, has a gap that no jurisdictional policy will close.

What a supervisor will test

In an examination touching cross-border tokenised activity, the questions are predictable and rarely about the foreign regime itself. Which rails and counterparties are eligible, on whose determination, and on what evidence. How instrument-level reserve and redemption terms are established where the home regime does not prescribe them uniformly. How on-ledger positions reconcile to books and records, at what cadence, and what happened the last time they did not. Which fourth-party dependencies sit in the settlement or evidence path and what the substitution plan is. Whether screening covers every path, and when that coverage was last tested. And what the board saw, when it saw it, and what it decided.

Every one of those is answerable today with existing artefacts, from the eligibility register, the third-party register, the reconciliation record and the board pack. None of them requires a view on legislation, in Washington or anywhere else.

How this maps to the Cabier estate

The obligations described here are not new modules. They sit against the conduct spine obligations on financial-crime controls and on fraud and illicit-flow assurance, against the settlement assurance work on rail eligibility and finality telemetry, and against the third-party workbench where fourth-party dependency mapping and concentration analysis live. The comparative determination itself is a substantial-similarity artefact and is produced by the same method used for any other cross-regime mapping.

Frequently asked questions

Is Russian digital-asset activity unregulated?

No. Federal Law No. 259-FZ established a digital financial assets framework in 2020, and a subsequent experimental legal regime narrowed and extended it for defined use cases under Bank of Russia oversight. The accurate description is a regime that exists and is supervised, but is not equivalent in design to MiCA, the MAS stablecoin standard or a US charter regime on reserve composition, redemption floors and public attestation cadence.

What does non-equivalent mean in practice for a control estate?

It means an institution cannot discharge its own obligation by pointing to the counterparty's home authorisation. Where a regime is equivalent, the home rulebook can carry part of the evidential weight. Where it is non-equivalent, the institution must generate the missing evidence itself or decline the exposure.

Is this article a sanctions analysis?

No, and deliberately so. It treats sanctions and export-control screening as a control category with its own evidential requirements. It makes no attribution claims about specific entities, transactions or evasion conduct, and it names no designated parties. Institutions should take designation and licensing questions to counsel and to primary regulator sources.

Why compare Russia to MiCA rather than to another emerging regime?

Because MiCA Titles III and IV are currently the most detailed published text in force on issuance, reserve composition, redemption and disclosure for fiat-referenced tokens. It is the most useful measuring stick available, whatever the jurisdiction being measured.

Does the experimental legal regime create a recognised cross-border settlement rail?

It creates a domestically supervised environment in which defined cross-border settlement use has been advanced. Recognition by other jurisdictions is a separate question and is not established by the experimental regime itself.

Can a rouble-referenced token satisfy MiCA reserve requirements?

Only if it were issued by an authorised issuer under MiCA and held reserves meeting the prescribed composition, segregation and custody rules. Compliance with a non-EU regime does not create MiCA compliance, and the reverse is equally true.

What is the difference between qualified-investor gating and a retail perimeter?

Qualified-investor gating restricts who may hold or trade an instrument, which reduces conduct and suitability exposure but does not reduce settlement, custody or reserve risk. Institutions sometimes read gating as a proxy for prudential quality. It is not.

How should a distributing institution treat a token whose reserve terms are set per issuance?

As a bespoke credit and liquidity exposure requiring instrument-level diligence, not as a fungible class exposure. Where terms vary by issuance decision, class-level policy is insufficient and each series needs its own control record.

Does a non-equivalent regime automatically mean higher risk?

Not automatically, and saying so is imprecise. It means the risk is less evidenced. Unevidenced risk is a governance defect regardless of the underlying quality of the arrangement.

What should the board see on this topic?

A rail and counterparty eligibility register showing which regimes the institution has assessed as equivalent, substantially similar or non-equivalent, the evidence supporting each determination, the date of assessment, and the exposures currently booked against each category.

How often should equivalence determinations be refreshed?

At least annually, and on trigger. Triggers include a change in the home regime's published rulebook, a change in the institution's exposure materiality, an incident at the counterparty or platform, and any change in designation or licensing status touching the arrangement.

Is substantial similarity the same as equivalence?

No. Equivalence is a determination made by a regulator about a regime. Substantial similarity, as Cabier uses it, is an institution's own attestation that a control designed against one named regime satisfies the operative requirement of another. It is a method for building once and evidencing several times; it is not a substitute for a regulator's recognition decision.

What role does settlement finality play here?

A central one. Where a token settles on a platform whose finality is defined by platform rules rather than by statute or a recognised system designation, the institution's own books-and-records reconciliation and dispute-resolution controls carry the weight that finality would otherwise carry.

Does this affect institutions with no Russian exposure?

Frequently yes, indirectly. Fourth-party dependencies — a platform operator, a custodian, an oracle or a data provider — can introduce exposure that the direct counterparty register does not show. That is a third-party risk mapping problem, not a jurisdiction problem.

What is the minimum defensible position for an institution that simply will not engage?

A documented determination not to engage, the scope of that determination, the screening controls that enforce it across onboarding, payments and custody paths, and evidence that the controls have been tested. A decision without an enforcement mechanism is not a control.

How does this interact with the pending US legislation?

Very little, which is the point. Cross-border eligibility questions are answered today by state charter conditions, prudential guidance, MiCA where in scope, and sanctions and AML obligations. The federal market-structure bill would change the label and the reporting format, not the diligence.

Where does this sit in the obligation taxonomy?

Primarily against the conduct spine obligations on financial-crime controls and illicit-flow assurance, and against the settlement assurance work on rail and counterparty eligibility. It also touches third-party risk where platform operators sit in the transaction path.

What would change our reading of the Russian regime?

A published reserve-composition standard with a mandated public attestation cadence and a statutory redemption floor, applied consistently across issuances rather than per issuance decision. Those three features are what separate a supervised market from an evidenced one.

Glossary

Digital financial asset (DFA)

The statutory category under Russian Federal Law No. 259-FZ covering rights recorded in a registered information system, including monetary-claim and security-like rights.

Experimental legal regime

A time-bound, scope-limited legal environment permitting defined activity under supervision, used to test arrangements outside the general regime.

Information-system operator

The registered entity operating the platform on which DFAs are issued and recorded; the practical locus of supervision.

Exchange operator

A registered entity permitted to organise trading in DFAs, distinct from the issuance platform operator.

Qualified investor gating

A restriction limiting holding or trading to investors meeting statutory sophistication criteria.

Fiat-referenced token

A token whose value references one or more official currencies; the regulatory category most regimes address first.

EMT (e-money token)

MiCA's category for a token referencing a single official currency, carrying the strictest reserve and redemption duties under Title IV.

ART (asset-referenced token)

MiCA's category for a token referencing a basket or non-single-currency value, governed under Title III.

Reserve composition

The specific permitted assets backing a token, their liquidity profile, custody arrangement and concentration limits.

Segregation

The requirement that reserve or client assets be held apart from the issuer's own estate so they survive its insolvency.

Redemption floor

A statutory minimum right to redeem at par within a defined period, independent of contract terms.

Attestation cadence

The mandated frequency of independent reporting on reserves, and whether that report is public or supervisory-only.

Equivalence

A regulator's determination that another jurisdiction's regime achieves comparable outcomes, permitting reliance.

Substantial similarity

An institution's own evidenced attestation that a control satisfies the operative requirement of a second named regime.

Non-equivalence

The condition of a regime that exists and is supervised but has not been determined comparable; evidence must be generated locally.

Rail eligibility

A documented determination that a specific settlement rail may be used for a specific exposure, with the controls that enforce it.

Settlement finality

The point at which a transfer becomes irrevocable, and the legal basis on which that irrevocability rests.

Books-and-records reconciliation

The control proving on-ledger positions agree to the institution's own accounting records at a defined frequency.

Fourth-party dependency

A material dependency of the institution's direct counterparty that sits in the transaction or evidence path.

Screening control

The operational mechanism enforcing a restriction across onboarding, payment, custody and distribution paths.

Illicit-flow assurance

The control set evidencing tracing, escalation and intervention capability over suspect on-ledger flows.

Wind-down plan

A tested plan for the orderly cessation of an arrangement, including reserve realisation and holder redemption.

Concentration limit

A ceiling on exposure to a single asset, counterparty, custodian or platform.

Designation status

The current listing position of a party under an applicable sanctions regime; a legal fact to be checked at source, not inferred.

Eligibility register

The board-visible record of assessed regimes, determinations, evidence, dates and booked exposures.

Sources and citations

Primary sources only. Designation, licensing and register positions change; verify at source before relying on any statement below.

1. 01

   Russian Federation, Federal Law No. 259-FZ of 31 July 2020, On Digital Financial Assets, Digital Currency and Amendments to Certain Legislative Acts

   Primary statutory basis for the DFA category and the registered-operator perimeter.

2. 02

   Bank of Russia — registers of information-system operators and DFA exchange operators

   The authoritative record of which platforms sit inside the supervised perimeter.

3. 03

   Bank of Russia — public materials on the experimental legal regime for digital-asset settlement

   Scope, participant eligibility and duration of the experimental environment.

4. 04

   Regulation (EU) 2023/1114 (MiCA), Titles III and IV

   Reserve composition, segregation, redemption-at-par and disclosure requirements for ARTs and EMTs.

5. 05

   European Banking Authority — MiCA technical standards and guidelines

   Detailed reserve, own-funds and reporting specifications.

6. 06

   Monetary Authority of Singapore — Stablecoin Regulatory Framework and Payment Services Act

   Single-currency stablecoin reserve, redemption-period and attestation standard.

7. 07

   Dubai Virtual Assets Regulatory Authority — Rulebooks, including the fiat-referenced virtual asset rulebook

   Licensing, reserve verification and disclosure requirements.

8. 08

   OCC, Federal Reserve and FDIC — interpretive letters and supervisory guidance on crypto-asset and tokenised-deposit activity

   Prudential expectations for US supervised institutions.

9. 09

   US state banking departments — limited-purpose trust charter conditions and money-transmitter requirements

   The operative US authorisation and examination path.

10. 10

    Congress.gov — H.R.3633 (119th Congress) legislative actions and Senate Legislative Calendar, General Orders No. 423

    Verified procedural status of the pending US market-structure bill.

11. 11

    FATF — Recommendation 15 and virtual-asset guidance, including the travel rule

    The international standard underlying counterparty and originator information duties.

12. 12

    OFAC, EU Council and UK OFSI consolidated lists and guidance

    Authoritative designation and licensing sources; to be checked at source for any live determination.