Reserve Requirements and Asset Backing

Regulator-Grade Guidance Document Last Updated: 4 June 2026

Executive Summary

Reserve requirements form the structural backbone of stablecoin regulation worldwide. The central question every jurisdiction must answer is deceptively simple: what stands behind the promise that one token equals one dollar (or euro, or yen)? The answers vary dramatically -- from Japan's original requirement that 100% of trust-type stablecoin backing be held in demand deposits, to Brazil's approach of embedding stablecoins within existing foreign exchange prudential rules rather than specifying reserve composition at all.

This document synthesises reserve, custody, audit, and asset-backing requirements across 18 jurisdictions, organised by regulatory model rather than by country. Every factual claim mirrors the verified per-jurisdiction fact matrix; where the matrix records that a point is not established in the underlying sources, the claim is marked NOT VERIFIED. The analysis concludes with a recommended tiered framework drawing on the strongest elements of each approach.

1. Global Landscape: Reserve Models

Stablecoin reserve regimes worldwide fall into five distinct architectural models. Understanding these models is essential before examining individual jurisdiction details.

1.1 Full-Reserve Prescriptive Model

Defining characteristic: Legislation or regulation specifies the asset classes, maturity limits, and (in some cases) allocation percentages that constitute permissible reserves. The regulator prescribes not just that reserves must exist but what they must contain.

Jurisdictions:

This is the dominant model globally and the one toward which regulation is converging. All jurisdictions in this category mandate a minimum 1:1 backing ratio.

1.2 Banking-Law Integration Model

Defining characteristic: Rather than creating bespoke stablecoin reserve rules, the jurisdiction classifies stablecoin issuance as deposit-taking (or functionally equivalent), bringing it under existing banking prudential requirements.

Jurisdictions:

Switzerland treats fiat-backed stablecoins as public deposits under its "same business, same risks, same rules" principle. Issuers must either hold a full banking licence (subject to Basel III capital and liquidity requirements), a FinTech licence (deposits up to CHF 100 million must remain uninvested -- effectively a 100% cash reserve earning no interest), or rely on a Swiss-bank default guarantee covering the total of all stablecoin-holder claims. Proposed FinIA amendments (consultation closed 6 February 2026, not yet enacted) would reserve "Regulated Stablecoin" issuance exclusively to a new Payment Institution licence category, with the FinIA dispatch to Parliament expected H2 2026 at the earliest.

1.3 Foreign Exchange Classification Model

Defining characteristic: Stablecoin transactions are classified as foreign exchange operations, embedding reserve and prudential oversight within the existing FX regulatory architecture.

Jurisdictions:

Brazil does not impose a standalone 1:1 statutory reserve-composition regime. Instead, fiat-referenced stablecoin transactions are classified as foreign-exchange operations under Resolution 521/2025, bringing them under existing FX prudential rules, and proof-of-reserves is verified through independent technical certification under Instruction 701/2026 (Art. 65 of Resolution 520). Mandatory client-asset segregation functions as a de facto reserve protection. Algorithmic stablecoins are banned outright.

1.4 CASP-Focused Model (No Issuance Regime)

Defining characteristic: The jurisdiction regulates the trading, custody, and distribution of stablecoins through service-provider licensing but does not regulate the issuance process or reserve composition.

Jurisdictions: Turkey (Law No. 7518 + CMB Communiques), South Africa (FSCA FAIS Declaration), India (no comprehensive framework)

These jurisdictions rely on foreign-issued stablecoins (primarily USDT and USDC) and address consumer protection through asset-segregation requirements at the exchange/custodian level rather than at the issuer level. Turkey's MASAK Circular No. 29 imposes stablecoin-specific transfer limits (USD 3,000/day, USD 50,000/month) but no reserve-backing requirements. South Africa's IFWG 2025 Stablecoin Landscape Diagnostic explicitly identified the absence of reserve verification as a critical regulatory gap.

1.5 Emerging/Proposed Model

Defining characteristic: Dedicated legislation is in progress -- in exposure draft, pending as a bill, or enacted but not yet in force.

Jurisdictions:

Australia's proposed Tranche 1 framework (exposure draft released 12 March 2026, consultation closed 9 April 2026, not enacted) mandates 1:1 backing for tokenised stored-value-facility/currency stablecoins, with APRA registration proposed for major issuers above an A$200 million threshold. South Korea's Digital Asset Basic Act (DABA) was introduced 10 June 2025 and re-proposed/consolidated by the ruling Democratic Party in April 2026, but remains a pending bill -- review has been pushed into H2 2026 amid an unresolved bank-vs-fintech issuer dispute. Nigeria's regulated naira stablecoin cNGN operates under SEC/CBN supervision, but Nigeria has no dedicated stablecoin issuance/reserve regime; its issuer was admitted to test under the SEC's Regulatory Incubation / Accelerated Regulatory Incubation Programme (RI/ARIP) track on 29 August 2024 (not granted full registration).

2. Comparative Analysis: Reserve Composition Rules

2.1 Qualifying Asset Classes

The table below maps eligible reserve assets across all jurisdictions with prescriptive requirements. The convergence on short-dated government securities and cash deposits is striking, but meaningful differences remain in maturity limits, credit-quality thresholds, and treatment of expanded asset classes.

2.2 Currency Denomination Requirements

A critical but often overlooked dimension of reserve policy: must reserves be denominated in the same currency as the stablecoin peg?

Singapore's approach is the most explicit: the SCS framework requires reserve assets to be denominated in the currency of the stablecoin peg. This prevents the currency-mismatch risk that contributed to fragility in earlier stablecoin arrangements where, for example, a USD-pegged stablecoin might be partially backed by assets denominated in other currencies.

2.3 Maturity Restrictions

Maturity limits on reserve assets serve as a proxy for liquidity-risk management. Shorter maturities mean reserves can be converted to cash more quickly to meet redemption demands, but at the cost of lower yield for the issuer.

Most restrictive: Singapore (3 months), Japan post-2025 amendment (3 months for the bond portion that may be invested), US GENIUS Act (93 days for T-bills), ADGM FRT (3 months for government debt)

Moderately restrictive: UAE CBUAE bank-subsidiary reserves (6 months for UAE govt bonds / Monetary Bills), Hong Kong (3 months for deposits, 1 year for debt securities), UK FCA (1 year for core eligible assets)

Least restrictive among prescriptive regimes: EU MiCA (not specified numerically; "highly liquid, minimal market risk")

The 3-month maximum adopted by Singapore and Japan reflects a deliberate policy judgment: reserves should be liquid enough to meet redemption demands under stress without forced sales at material discounts. The US GENIUS Act's 93-day limit for Treasury bills is functionally equivalent.

3. Segregation and Custody Requirements

3.1 Third-Party Custodian Mandates

Every jurisdiction with a prescriptive reserve framework requires some form of asset segregation. The specifics vary significantly.

Statutory trust / trust arrangements required:

• UK (FCA): Backing assets held in a statutory trust; segregated by stablecoin product (not pooled across products); minimum 5% in on-demand deposits
• Canada (Stablecoin Act): Reserves held by qualified custodians (federally regulated trust companies and Schedule I banks with Bank of Canada approval); segregated from both custodian and issuer assets; protected from creditor claims in bankruptcy
• Bahrain: Segregated accounts with an authorised third party (bank/investment firm/custodian) under written contract; no commingling with operating funds

Segregation with institutional custodian requirements:

• Singapore: Approved custodians must be either (a) financial institutions licensed for custodial services in Singapore, or (b) overseas custodians with a minimum credit rating of "A-" with a MAS-regulated Singapore branch
• Hong Kong: Effective trust arrangements required; reserve assets segregated from the issuer's proprietary assets; tokenised reserves permitted; partial overseas custody subject to HKMA approval
• UAE: Segregated escrow accounts at UAE-licensed banks (CBUAE/VARA); ADGM requires separate reserve pools per FRT type; reserves unencumbered and not pledged

Banking-law segregation:

• Japan: Trust structure provides bankruptcy-remote protection for trust-type stablecoins; fund-transfer-type stablecoins require strict segregation of client funds from the provider's treasury
• Switzerland: The DLT Act provides explicit segregation rules for crypto-assets in custody during bankruptcy, requiring individual attribution and continuous beneficial-owner control; the default-guarantee arrangement gives each customer an individual claim against the guaranteeing bank

Federal standards with multiple custodian options:

• US (GENIUS Act): Reserves must be segregated from operational funds; specific custodian requirements being elaborated through OCC, FDIC, and Federal Reserve rulemaking (OCC NPRM issued 25 February 2026; FDIC prudential NPRM approved 7 April 2026)
• EU (MiCA): Reserve assets are legally and operationally segregated from the issuer's estate and from other token reserves (Art. 37); the custodian must be a separate legal entity; reserves placed within 5 working days

Provider-level segregation only (no issuer reserve regime):

• Brazil: Mandatory client-asset segregation from proprietary assets; firms prohibited from using client assets
• Turkey: Customer crypto-assets segregated from CASP operational assets; protected from creditor and bankruptcy claims (at the CASP level, not the issuer level)
• Nigeria: Reserve assets placed with SEC-approved digital-asset custodians; cNGN reserves held in commercial banks under SEC supervision
• South Africa: Not regulated for issuers -- a gap identified by the IFWG Diagnostic

3.2 Bankruptcy Remoteness

Bankruptcy remoteness -- the legal assurance that reserve assets remain available for holder redemption even if the issuer becomes insolvent -- is the most critical structural protection. Approaches differ:

• Bermuda: The BMA's SCPS Guidance requires backing assets to be held in structures that are bankruptcy-remote from group entities, issuers, managers, or any parties involved in token issuance
• Canada (Stablecoin Act): Reserve assets may not be pledged, encumbered, or used for anything other than redemptions; protected from creditor claims in bankruptcy
• EU (MiCA): Reserve assets are used exclusively for the benefit of token-holders and are legally shielded from the issuer's general creditors in insolvency (Art. 37)
• UAE (CBUAE): Legal ring-fencing gives consumers priority claims on reserves in insolvency
• Japan (Trust-type): Trust structure inherently provides bankruptcy-remote segregation; the 2025 amendment adds FSA domestic-asset-holding orders to strengthen recovery for Japanese users
• Switzerland: Banking-law depositor preference plus the DLT Act's explicit segregation of custodied crypto-assets; the default guarantee gives each customer an individual claim against the bank in issuer bankruptcy
• Turkey: Customer crypto-assets "cannot be seized by the service provider's creditors; they cannot be included in the assets to be distributed in the event of bankruptcy" (Law No. 7518) -- but this applies at the CASP level, not the stablecoin-issuer level

For several jurisdictions -- including the US, Singapore, Hong Kong, Bahrain, Brazil, Nigeria, South Africa, South Korea, and Australia -- the matrix records insolvency-priority treatment as NOT VERIFIED or not specified, even where structural segregation or a direct legal claim is required.

3.3 Multi-Custodian and Concentration Requirements

Few jurisdictions explicitly require diversification across multiple custodians. This represents a regulatory gap, as concentration of reserves with a single custodian creates single-point-of-failure risk.

• EU (MiCA): Requires highly liquid instruments with minimal concentration risk and diversification of reserve assets to avoid excessive concentration (Titles III/IV reserve rules)
• UK (FCA): The statutory-trust backing framework requires issuers to model redemption scenarios and maintain appropriate liquidity buffers, implicitly encouraging diversification
• Singapore: Does not mandate multi-custodian arrangements but requires risk-management policies covering concentration risk

4. Audit and Attestation Requirements

4.1 Full Audit vs. Attestation

The distinction between a full audit and an attestation is material. An attestation is a point-in-time examination of specific management assertions (e.g., "reserves exceeded outstanding tokens on date X"). A full audit provides a comprehensive opinion on whether financial statements are fairly presented. Attestation is cheaper and faster but provides narrower assurance.

4.2 The Tether Problem: Attestation vs. Audit in Practice

The case of Tether (USDT) illustrates the limits of attestation-only regimes. Tether has historically relied on periodic attestation reports rather than submitting to a full, independent audit of its reserves. Attestations verify reserve adequacy at specific points in time but do not provide the continuous, comprehensive assurance that a full audit delivers.

Tether has held substantial reserve exposure to US Treasuries alongside asset classes -- such as gold, bitcoin, and secured loans -- that would not qualify as permitted reserves under the GENIUS Act, MiCA, or Singapore's SCS framework. To serve US persons under the GENIUS Act framework, Tether launched USAT on 27 January 2026, a GENIUS-compliant payment stablecoin issued via Anchorage Digital Bank, coexisting with the globally circulating USDT.

The CFTC's 2021 enforcement action against Tether found that the company had made "untrue or misleading statements and omissions of material fact" regarding reserve backing. Specifically, Tether represented that USDT was backed 1:1 by US dollars held in reserve, but the CFTC found that for 849 of the 1,096 days during a 26-month period (June 2016 to February 2019), USDT was not fully backed. Tether paid $41 million in civil penalties.

This enforcement history directly informed the GENIUS Act's requirement for monthly executive-certified reserve reports and annual audits for issuers exceeding $50 billion -- provisions clearly targeted at ensuring that the Tether experience is not repeated. Related enforcement followed the Terra/UST episode: in December 2024 the SEC settled with Tai Mo Shan (Jump) for $123 million over UST manipulation.

5. Reserve Yield and Interest Prohibition

5.1 The Core Policy Question

When a stablecoin issuer holds billions in Treasury bills as reserves, those assets generate yield. The policy question is: who captures that yield, and can any of it be passed to token holders?

This question sits at the intersection of securities law (yield-bearing instruments may be securities), monetary policy (interest-bearing stablecoins could disintermediate bank deposits), and consumer protection (should holders benefit from the reserves they effectively fund?).

5.2 Interest Prohibition Jurisdictions

EU (MiCA): Articles 40 and 50 prohibit ART and EMT issuers (and CASPs) from granting interest or any benefit related to the length of time a holder holds the token. This is the strictest prohibition globally. The ART/EMT provisions have applied since 30 June 2024 and have caused major crypto platforms to disable stablecoin reward programmes for European users.

Hong Kong: The Stablecoins Ordinance (Cap. 656) prohibits licensees from paying any interest, profit, or other return to holders of their stablecoins, consistent with the MiCA approach.

US (GENIUS Act): A payment-stablecoin issuer may not pay interest or yield to stablecoin holders. This positions payment stablecoins firmly as payment instruments rather than investment products, and preserves the SEC's April 2025 staff statement that "Covered Stablecoins" are not securities.

Canada (Stablecoin Act): Issuers are prohibited from granting any form of interest, yield, rewards, or other inducement to stablecoin holders.

5.3 Jurisdictions Permitting Yield

Bahrain: The SIO Module explicitly permits yield-bearing stablecoins, including Sharia-compliant reward-bearing instruments. This is a notable outlier among jurisdictions with comprehensive stablecoin frameworks.

UAE (ADGM): Yield-bearing FRTs are permitted if reserves exceed par value, provided they are not marketed as an investment or savings product.

Switzerland: No prohibition on yield distribution. The FinTech licence requires that deposits remain uninvested and earn no interest, but this applies to the licensing pathway, not as a prohibition on the stablecoin product itself; multiple CHF stablecoin projects are operational (DCHF, tCHF, ZCHF).

5.4 Issuer Retention of Reserve Yield

In jurisdictions that prohibit interest to holders, the issuer retains all yield generated by reserve assets, creating substantial revenue streams from reserves held primarily in short-dated government securities.

The policy rationale for permitting issuers to retain yield while prohibiting pass-through to holders is that: (a) it preserves the distinction between payment instruments and investment products, (b) it avoids securities classification, and (c) it funds the issuer's compliance, operational, and capital costs. Critics argue this transfers wealth from holders to issuers without adequate justification.

5.5 Tax Treatment of Reserve Yield

The tax treatment of yield earned on reserves held by stablecoin issuers is governed by the general corporate income-tax rules of the jurisdiction where the issuer is resident. No jurisdiction has enacted stablecoin-specific tax provisions for reserve yield. In zero-tax jurisdictions like Bermuda (no corporate income tax, no capital gains tax) and Bahrain (no corporate or capital gains tax), reserve yield is effectively untaxed at the entity level; in the US, an issuer's reserve yield is subject to federal and state corporate income tax.

6. The Terra/LUNA Collapse and Regulatory Responses

6.1 Verified Timeline

The collapse of the Terra/LUNA algorithmic stablecoin ecosystem in May 2022 was the single most consequential event in stablecoin regulatory history, directly precipitating or accelerating legislation across many jurisdictions.

• 7 May 2022: Two large addresses withdrew 375 million UST from the Anchor lending protocol; an 85-million UST swap for USDC triggered cascading sell pressure. UST fell from $1.00 to $0.985.
• 9 May 2022: UST de-pegged, falling to approximately $0.35. The algorithmic stabilisation mechanism -- which allowed 1 UST to be redeemed for $1 worth of LUNA -- triggered hyperinflationary LUNA minting as holders rushed to exit.
• 10-13 May 2022: LUNA's supply ballooned from billions to trillions of tokens; its price collapsed from $87 to less than $0.00005. UST fell below $0.20.
• 13 May 2022: The Terra blockchain was temporarily halted. Approximately $45 billion in combined market capitalisation was destroyed within one week.
• 2024-2025: A jury found Terraform Labs and Do Kwon liable for securities fraud (April 2024); a court approved a $4.47 billion Terraform settlement with the SEC (June 2024); Do Kwon was sentenced to 15 years in US federal prison (SDNY, 11 December 2025).

6.2 Regulatory Responses by Jurisdiction

The Terra collapse catalysed the following specific regulatory actions:

Explicit algorithmic stablecoin bans or restrictions:

• UAE (CBUAE PTSR): Algorithmic stablecoins and privacy tokens strictly prohibited from issuance, service provision, or promotion across CBUAE, ADGM, and DFSA (PTSR effective 6 July 2024)
• Brazil (BCB Resolution 520/521): Algorithmic stablecoins banned -- Brazilian VASPs may not offer or intermediate purely algorithmic pegs
• Singapore: Algorithmic stablecoins excluded from the "MAS-regulated stablecoin" SCS designation; they remain regulated as DPTs without the privileged label
• US (GENIUS Act): Payment stablecoins are defined to require asset backing, effectively excluding algorithmic/endogenously-collateralised models from PPSI status (they remain outside the framework, potentially within SEC/CFTC jurisdiction)
• South Korea: The Terra collapse originated from Korean-founded Terraform Labs and directly catalysed the VAUPA (in force July 2024) and the pending DABA; the DABA would require 100%+ reserve backing, implicitly excluding algorithmic models

Reserve-requirement tightening directly linked to the collapse:

• Japan: The PSA Amendment Act 2022 (effective 1 June 2023), creating the EPI framework, was enacted in the wake of the Terra collapse. The framework's insistence on par-value redemption and reserve backing was a direct response.
• EU: While MiCA was already under development, the Terra collapse accelerated political will; Recital 41 excludes pure supply-adjusting algorithmic tokens from the ART category.

6.3 Why Terra Failed: Structural Analysis for Regulators

The Terra/LUNA collapse was not a failure of execution but a failure of mechanism design. Understanding the structural cause is essential for regulators drafting reserve requirements.

UST was an endogenously collateralised stablecoin: its value was maintained by an arbitrage mechanism linking it to LUNA, a separate token within the same ecosystem. When UST holders lost confidence, they redeemed UST for newly minted LUNA, expanding LUNA's supply and crashing its price. This destroyed the collateral that was supposed to maintain UST's peg -- a reflexive death spiral with no external reserve to absorb losses.

The lesson: reserves must be exogenous to the stablecoin system. Cash, government securities, and bank deposits held in segregated accounts with regulated custodians cannot enter a reflexive spiral because their value is independent of the stablecoin's market dynamics. Every major prescriptive jurisdiction has now embodied this principle in its framework.

7. Rehypothecation and Lending of Reserves

7.1 The Risk

Rehypothecation -- the practice of using customer-pledged collateral to secure the custodian's or issuer's own borrowing -- creates hidden leverage and counterparty risk in reserve pools. If reserve assets are lent out or pledged as collateral, they may not be immediately available for redemption, and their loss through counterparty default would leave the stablecoin under-collateralised.

7.2 Jurisdictions Explicitly Prohibiting Rehypothecation

US (GENIUS Act): Reserves may not be rehypothecated or pledged except in three narrow cases: (1) to satisfy margin obligations in connection with investments in permitted reserves, (2) to satisfy obligations associated with standard custodial services, or (3) to create liquidity to meet reasonable expectations of redemption requests. These exceptions are narrowly drawn and closely track the operational necessities of reserve management.

EU (MiCA): Issuers may not lend ART reserves, and reserve assets must be legally and operationally segregated and maintained solely for the benefit of token-holders.

Canada (Stablecoin Act): Reserve assets "may not be pledged, encumbered, or used for anything other than redemptions."

Singapore: The SCS framework prohibits issuers from lending, staking, or dealing in digital payment tokens other than the SCS being issued. While not framed as a "rehypothecation prohibition," the business-restriction rules achieve the same effect by preventing any use of reserves beyond their custodial function.

UAE: Reserve assets must be unencumbered and not pledged (CBUAE/VARA/ADGM).

7.3 Jurisdictions Without Explicit Prohibition

Hong Kong, Japan, Bahrain, Bermuda, Switzerland, UK: These jurisdictions either do not explicitly address rehypothecation of stablecoin reserves in their published frameworks or address it implicitly through segregation and custody requirements. The risk remains that creative financial structuring could circumvent segregation requirements in jurisdictions without explicit rehypothecation bans.

7.4 Recommendation

Explicit prohibition of rehypothecation, with narrowly drawn exceptions for operational necessities (as in the GENIUS Act model), should be a baseline requirement. Segregation requirements alone are necessary but not sufficient -- they address where reserves are held but not what is done with them.

8. Comparative Jurisdiction Table

The following table provides an at-a-glance comparison across all covered jurisdictions. It is designed as a reference tool for regulators conducting gap analysis against their own frameworks.

9. Recommended Framework

Based on the comparative analysis, the following tiered framework represents regulatory best practice as of June 2026. It draws from the strongest elements across all examined jurisdictions.

Tier 1: Systemic Stablecoins (>$10 billion outstanding)

Tier 2: Material Stablecoins ($1 billion - $10 billion outstanding)

Tier 3: Small Stablecoins (<$1 billion outstanding)

10. Decision Matrix for Regulators

The following matrix maps key policy decisions that every jurisdiction must make when designing a stablecoin reserve framework.

Jurisdictions adopting the "conservative" column across all dimensions approximate the Singapore/UK BoE model. Those adopting "moderate" across the board approximate the GENIUS Act model. The "permissive" column generally characterises principles-based jurisdictions like Bermuda or pre-reform Switzerland.

Verification Notes

This document is a refresh synthesis: every claim mirrors the canonical per-jurisdiction fact matrix (.claude/stablecoin-fact-matrix.json), which is the verified source of truth for this domain. No web searches were conducted for this pass. Where the matrix records a point as NOT VERIFIED or not established in the underlying sources, this document reflects that honestly rather than asserting a value.

Corrections Applied This Pass (2026-06-04)

• Hong Kong: Stablecoins Ordinance is Cap. 656 (corrected from any prior "Cap. 636" reference); in force 1 Aug 2025; first two issuer licences (Anchorpoint FRS01, HSBC FRS02) granted 10 Apr 2026.
• Canada: The Stablecoin Act is enacted via Bill C-15 (Royal Assent 26 Mar 2026) but not yet in force (in-force date set by Order in Council; draft regulations not yet published as of 4 Jun 2026). Prior "proposed Stablecoin Act / in Parliament" framing corrected.
• Japan: The PSA Amendment Act 2025 takes effect on a single date, 1 June 2026 (per the FSA Cabinet Order published 22 May 2026). Any prior "13 June operational deadline" framing removed; 13 June 2025 was the promulgation date of the amendment, not an effective date.
• United States: GENIUS Act implementing NPRMs advanced Apr-May 2026 -- OCC NPRM (25 Feb 2026), FDIC prudential NPRM (7 Apr 2026), Treasury (FinCEN+OFAC) joint AML/CFT and sanctions NPRM (8 Apr 2026), FDIC illicit-finance NPRM (22 May 2026). Tether launched USAT (GENIUS-compliant, via Anchorage Digital Bank) on 27 Jan 2026. Effective date is the earlier of 18 Jan 2027 or 120 days after final rules.
• South Korea: DABA is a pending bill (introduced 10 Jun 2025, re-proposed/consolidated by the Democratic Party Apr 2026), not enacted; review pushed into H2 2026. Reserve, audit, and redemption provisions are proposed, not in force.
• Nigeria: Integrated into all tables this pass. The cNGN issuer was admitted to test under the SEC's RI/ARIP track on 29 Aug 2024 (not granted full registration / AIP). Nigeria has no dedicated stablecoin issuance/reserve regime; cNGN is 1:1 to the Naira with reserves held in commercial banks under SEC supervision and SEC-approved custodians.

Aggregate Derivations (recomputed from the matrix)

• Jurisdictions covered: 18 (United States, Bermuda, Brazil, Canada, Australia, Hong Kong, India, Japan, Singapore, South Korea, EU, Switzerland, Turkey, UK, Bahrain, Nigeria, South Africa, UAE).
• Mandating a 1:1 / 100% backing ratio (enacted or proposed): US, EU (ART & EMT), Singapore, Japan (fund-transfer/trust), Hong Kong, UK, Canada (enacted, not in force), Switzerland (effectively, via banking/FinTech/guarantee), Bermuda, Bahrain, UAE, Australia (proposed), South Korea (proposed). Brazil uses proof-of-reserves under FX classification rather than a standalone statutory ratio; Turkey, South Africa, and India impose no issuer reserve ratio.
• Explicit rehypothecation / encumbrance prohibition: US, EU, Canada, Singapore (via business restrictions), UAE.
• Interest to holders prohibited: US, EU (ART & EMT), Hong Kong, Canada, UAE (CBUAE). Yield permitted: Bahrain, UAE (ADGM FRT, conditional), Switzerland (no prohibition).
• Algorithmic stablecoins banned outright: UAE (UAE-wide), Brazil. Excluded-from-framework (effectively inoperable, not a named ban): US, EU, Singapore, Japan, Hong Kong, UK, Canada, South Korea (proposed). Not specifically addressed: Switzerland, Bermuda, Nigeria, Turkey, South Africa, India, Australia.

Items Marked NOT VERIFIED in the Matrix (carried through honestly)

• Brazil: Permitted reserve-asset composition is not prescribed in public BCB materials (NOT VERIFIED); attestation cadence NOT VERIFIED.
• Switzerland: No dedicated stablecoin reserve-attestation cadence (banking-law audit applies); algorithmic treatment not specifically addressed.
• Nigeria: Detailed eligible-asset composition beyond "reserves held in commercial banks" is NOT VERIFIED; cNGN flagged for absence of continuous proof-of-reserve.
• South Korea: Audit cadence and substance requirements NOT VERIFIED in enacted text (the DABA bill is not passed).
• Australia: Capital figures and precise audit cadence NOT VERIFIED in primary statutory text (Tranche 1 in exposure draft).
• Bermuda / Singapore / US / Hong Kong / Bahrain: Insolvency-priority treatment recorded as NOT VERIFIED or not specified in several files, even where structural segregation or a direct legal claim is required.
• UAE: CBUAE eligible-asset specifics beyond the published categories may be communicated through individual licensing conditions.