## crypto-assets-cg

## Source details

**Canonical URL:** [crypto-assets-cg](https://www.imf.org/-/media/files/data/statistics/bmp7/implementation-support/crypto-assets-cg.pdf)

## Other formats

- [Markdown version](/-/media/files/data/statistics/bmp7/implementation-support/crypto-assets-cg.pdf.md)
- [Structured JSON version](/-/media/files/data/statistics/bmp7/implementation-support/crypto-assets-cg.pdf.json)

---

### Purpose, scope, and organization of the Guide
- The Guide states its primary purpose is to offer practical guidance on source data and compilation methods for measuring crypto asset-related transactions and positions in macroeconomic statistics, consistent with the System of National Accounts 2025 (2025 SNA) and the Integrated Balance of Payments and International Investment Position Manual, seventh edition (BPM7).
- The Guide supports implementation of the 2025 SNA/BPM7 treatment of crypto assets and the G20 Data Gaps Initiative (DGI-3) Recommendation 11 on digital money (including central bank digital currencies, stablecoins, and other crypto assets used as means of payment).
- National compilers are encouraged to adapt recommended data sources and methods to national circumstances to produce consistent and comparable estimates across economies.
- Chapter organization:
  - Chapter I: Overview of crypto assets ecosystem (Purpose, Background, Current State of Play, Regulatory/Legal Frameworks, Organization of Guide).
  - Chapter 2: Methodological framework for recording crypto assets (typologies, issuance, other changes, exchanges/platforms, lending/borrowing, NFTs).
  - Chapter 3: Data sources and compilation methods (administrative, financial reports, surveys, commercial sources, NFTs, DGI-3 templates, data sharing).
  - Chapter 4: Country experiences and international initiatives; glossary and annexes provide examples, survey templates, and numerical recording examples.

### Current state of play — salient facts and systemic implications
- Market size and composition:
  - Number of crypto assets expanded from a few hundred (including Bitcoin) to nearly 20,000.
  - Total market capitalization stands at approximately $2  .5 trillion.
  - Bitcoin represents around 60 percent of that value.
  - Market peak and volatility: market reached a peak of $3 trillion in November 2021 and subsequently dropped below $1 trillion.
- Systemic risk considerations:
  - Crypto assets are not currently a major part of the global financial system but can be a potential source of systemic risk in certain jurisdictions.
  - Higher adoption and weak regulatory frameworks—especially in emerging and low-income economies—can elevate macro-financial and systemic risks through currency substitution, capital outflows, and interlinkages with the financial system.
- Macro-financial uses and risks:
  - Uses: cross-border payments including remittances; potentially faster and cheaper transactions.
  - Risks: money laundering, tax avoidance, potential impacts on monetary policy transmission if adopted as means of payment.
  - Regulatory uncertainty remains a significant challenge limiting broader integration.

### Definitions, typologies, and classification (2025 SNA/BPM7)
- Core definitions:
  - Crypto assets: digital representations of value that use cryptography and distributed ledger technology (DLT).
  - Fungible crypto assets: divisible and not unique (examples: Bitcoin, Ether, USD Coin, Tether).
  - Nonfungible crypto assets (NFTs): unique and nondivisible (examples: Bored Ape Yacht Club, CryptoPunks, NBA Top Shot).
  - Native crypto assets: originate on their own blockchain (examples: Bitcoin and Ether).
  - Non-native crypto assets: represent underlying financial or real-world assets via tokenization (examples: tokenized equities, bonds, real estate; crypto securities).
- 2025 SNA/BPM7 classification highlights:
  - Fungible crypto assets divided into:
    - crypto assets designed to act as a general medium of exchange;
    - crypto assets that only act as a medium of exchange within a platform or network;
    - security crypto assets.
  - Crypto assets with corresponding liabilities:
    - Issued by a monetary authority (e.g., qualifying CBDCs) — AF21 (currency and deposits).
    - Not issued/authorized by monetary authorities (e.g., stablecoins with a claim on the issuer) — AF22 (separate category under currency and deposits).
    - Payment tokens with corresponding liability — AF33 (short-term) and AF34 (long-term) debt securities subcategories.
    - Debt security crypto assets — AF3; Utility tokens — AF35; Equity crypto assets — AF51; Derivative crypto assets — AF71.
  - Crypto assets without corresponding liabilities designed as media of exchange (CAWLM and CAWLP) — classified as nonproduced nonfinancial assets and recorded under NF22 in capital account and balance sheet.
- Reserve asset treatment (BPM7 guidance):
  - Crypto assets without corresponding liabilities (e.g., Bitcoin, Ether) should not be included in reserve assets.
  - For crypto assets treated as financial assets to qualify as reserve assets, they must satisfy BPM7 paragraphs 6.69–6.80 criteria, including effective control by monetary authorities, being actual high-quality external claims on nonresidents, and being denominated and settled in convertible foreign currency.
  - The Guide notes most stablecoins currently appear not to meet these requirements, but a case-by-case assessment is necessary.

### Recording conventions for issuance, mining, validation, staking, and lending
- How fungible crypto assets enter circulation:
  - Miners validating transactions in crypto assets without corresponding liabilities receive explicit validation fees and implicit fees (block rewards in new crypto assets).
  - Non-mineable crypto assets may be released via explicit sale and/or as payment to validators in PoS or PoA systems.
- Treatment of mining/validation services:
  - Activities that create new crypto assets without corresponding liabilities are regarded as production activities.
  - Services of miners/validators measured as sum of explicit validation fees and implicit fees (newly released crypto assets).
  - Implicit fees are assumed collectively paid by existing holders; by convention, implicit fees are recorded as a cross-border transaction payable by owners of existing crypto assets to the producer, with the new crypto assets diluting the value of existing crypto assets.
  - The convention may create global asymmetries but promotes consistent recording.
- Staking treatment:
  - Active staking (direct via blockchain): accrued revenue (block rewards and explicit fees) treated as output of validation services and recorded as computer services in the balance of payments.
  - Passive staking (through CEX/DEX): staking revenue recorded on a gross basis; platform fees recorded as intermediate consumption for staking-as-a-service and classified as computer services.
  - Implicit fee convention parallels PoW treatment.
- Crypto lending:
  - Lending crypto assets without corresponding liabilities: fees payable to owners recorded as interest (on the lines of securities lending and gold loans); lenders remain economic owners (no capital account transaction recorded).
  - Lending of crypto assets with corresponding liabilities: treated as financial instruments; securities-lending treatment applies; if lender remains economic owner no transaction recorded in financial account.
  - Lending of stablecoins and CBDCs on blockchain: recorded along lines of cash lending; associated lending revenue recorded as interest.

### Numerical examples preserved exactly as reported
- Mining/validation numerical example (Economy A and B; market value per crypto asset USD 1,000):
  - Economy B institutional units pay transaction fee of 5 in crypto assets.
  - Economy A mining company receives 20 units of crypto assets as implicit fee.
  - Mining company incurs 18 crypto units on electricity and other inputs.
  - Entries recorded in Economy A BOP (in USD):
    - Services — Computer services (explicit fee—from Economy B) 5,000
    - Services — Computer services (implicit fee—from rest of the world) 20,000
    - Capital account — Acquisition/disposal of non-produced nonfinancial assets: 25,000 (5,000+20,000)
  - Economy A National Accounts (in USD):
    - Output (explicit fee + implicit fee) 25,000
    - Intermediate consumption 18,000
    - Gross value added 7,000
    - Change in assets — Crypto assets without corresponding liabilities 25,000
- Staking numerical example (Box 2.3 — Stabex stablecoin; 1 Stabex = 1 USD):
  - Stabex placed: 1000 units
  - Staking revenue earned by Cryptostake: 150 units
  - Fee retained by Cryptostake: 30 units
  - Payment to Institutional Unit A: 120 units
  - Intermediate inputs paid by Cryptostake domestically: 5 units
  - Economy A BOP current account balance (passive staker perspective): 120
  - Economy B BOP current account balance (validator perspective): 30
  - Economy A national accounts: Output 150; Intermediate consumption 30; Gross value added 120; Financial account 120
  - Economy B national accounts: Output 30; Intermediate consumption 5; Gross value added 25; Financial account 25
- Crypto lending numerical example (Box 2.8):
  - Institutional unit A lends 100 Bitcoins to Institutional unit B and receives principal + 5 Bitcoins as revenue after one year.
  - Institutional unit A lends 10,000 units of Tether to a unit in Economy C and receives 500 Tether as interest payment for one year.
  - Prices: one Bitcoin = 1,000 USD; one Tether = 1 USD.
  - Entries for Economy A — Bitcoin transaction (USD):
    - Current account — Earned Income — Interest: 5,000
    - Capital account — Acquisitions/disposals of nonproduced nonfinancial assets — Crypto assets without corresponding liability: 5,000
  - Entries for Economy A — Tether transaction (USD):
    - Current account — Earned Income — Interest: 500
    - Financial account: Other investment — Currency and deposits (Tether) -10,000; Currency and deposits (Tether) 500; Loans (Tether) 10,000
    - National accounts: Allocation of Earned Income Account — Interest: 500; Currency and deposits (Tether) -9,500 (-10,000+500); Loans (Tether) 10,000

### NFTs — classification and recording by type
- Definition and lifecycle:
  - NFTs: digital records on a blockchain associated with a digital or physical asset, each with a unique identifier and metadata; ownership history publicly recorded.
  - Minting typically via execution of a smart contract; payments for NFTs usually in the native fungible crypto asset of the hosting blockchain.
- Three NFT types (2025 SNA/BPM7):
  - First type: convey no ownership rights and only allow personal use — recorded as final consumption expenditure.
  - Second type: convey limited ownership rights beyond personal use — treated as nonproduced nonfinancial assets (contracts, leases and licenses) and recorded in the capital account if they confer realizable economic benefits.
  - Third type: convey full ownership rights — NFT is a digital recording of ownership; underlying asset already recorded in national accounts; purchase recorded as purchase of the underlying asset.
- Numerical NFT example (Box 2.10 — CryptoPunk #4567 sale):
  - NFT valued at USD 300,000. Seller pays fee of 0.5% to marketplace (Yuga Labs). Payment in Ether.
  - BOP entries for Economy A (USD):
    - Current account: -1,500 — Services — Nonfinancial intermediation services (0.5%*300,000 = 1,500)
    - Capital account: 1,500 — Acquisitions/disposals of nonproduced nonfinancial assets — Contracts, leases, and licenses (NFTs-second type) vis-à-vis Economy B: 300,000
    - Crypto assets without corresponding liability (Ether) vis-à-vis Economies B (300,000) and C (1,500): 1,500 300,000
  - Outcome: Seller receives $298,500 after paying $1,500 fee to platform in Economy C.

### On-chain data, consensus mechanisms, miners/validators, and stablecoin/NFT issuers
- On-chain data (Bitcoin example) provide:
  - Public address activity (balances, incoming/outgoing transactions), transaction details (transaction ID, pseudonymous addresses, crypto amounts, fees, timestamps), block data (block hash, height, miner address, transactions list, timestamp, size, nonce), and network data (total number of nodes, total transactions processed, hash rate).
- Consensus mechanisms:
  - Proof-of-Work (PoW): miners solve puzzles; known for high energy consumption; examples: Bitcoin and Ether (before its transition to PoS).
  - Proof-of-Stake (PoS): validators chosen based on staked coins; more energy efficient; examples: Ether (after transition) and Cardano.
  - Private blockchains: use mechanisms prioritizing efficiency (Proof of Authority, PBFT). PBFT requires a minimum of two-thirds of nodes to agree.
- Miners/validators incentives:
  - PoW miners earn explicit transaction fees and block rewards in newly minted crypto assets.
  - PoS validators earn staking rewards and transaction fees; staked tokens can be slashed for malicious behavior.
- Stablecoin issuers:
  - Issuers can be deposit-taking corporations or nonbank issuers (examples cited: Tether Limited, Circle; JPMorgan’s JPM Coin as bank-issued).
  - Designs: reserves backing (cash, deposits, short-term securities, sometimes nonfinancial assets), algorithmic stablecoins, hybrid approaches.
  - Examples: Tether (USDT), USD Coin (USDC), Binance USD (BUSD).
- NFT issuers:
  - Examples: Yuga Labs (Bored Ape Yacht Club), Nike (virtual sneakers), Decentraland (virtual land).

### Data sources and compilation methods (Chapter 3) — types, strengths, and limitations
- Covered source types:
  - Administrative and supervisory data (registries, regulatory reports, ITRS, tax records, seizure data).
  - Financial statements of mining companies, stablecoin issuers, crypto exchanges/custodians.
  - Alternative commercial and blockchain-based data (on-chain analytics firms).
  - Surveys of enterprises and households; targeted questionnaires for custodians and stablecoin issuers.
  - International data sharing and DGI-3 Recommendation 11 templates.
- Administrative sources — highlights and limitations:
  - CASP registries: regulators publish lists; registration and AML/CFT obligations can provide foundational data.
  - Regulatory reporting: useful if tailored for statistical reporting; common limitations include lack of explicit distinction between assets with/without corresponding liabilities, missing residency and sectoral holder data, coverage gaps for decentralized entities, confidentiality restrictions, and timeliness issues.
  - Tax data: useful for flows (realized gains/losses, declared income from mining/staking) but generally do not provide wallet balances or comprehensive stock data; frequency typically annual.
  - ITRS: captures cross-border crypto-related payments processed through domestic banking system; limitations include coverage only of bank-intermediated transactions and lack of asset position data.
  - Loss/theft/seizures: data compiled by law enforcement and private analytics firms; treated as other changes in volume (not transactions).
- Financial reports:
  - Mining companies: revenue components — block rewards and transaction fees; main intermediate input electricity; company financials often consolidated, lack residency breakdowns, limited revaluation disclosure.
  - Stablecoin issuers: financial statements may disclose reserve composition (examples include U.S. Treasury bills, repos, money market funds, cash, Bitcoins, precious metals); classification challenges for reserves that are nonfinancial (e.g., Bitcoins, precious metals).
  - Crypto exchanges/custodians: disclose assets under custody (AUC) and revenue lines (custody revenue, transaction revenue, staking revenue); financial statements often lack residency and sectoral breakdowns of holders and do not separate transactions from revaluations.
- Commercial on-chain analytics:
  - Firms (e.g., Chainalysis, Elliptic, Glassnode) provide market capitalization, on-chain transactions, wallet balances, entity classification heuristics, and geographic estimates.
  - Limitations: on-chain data capture only part of activity (off-chain CEX activity omitted), addresses do not reveal ultimate beneficial owners, internal transfers can be misinterpreted as economic flows, residency inference approaches rely on proxies, and commercial methodologies may not meet BPM7/2025 SNA standards.
  - Coverage example: Chainalysis covers about 90% of daily market volume across 120 digital assets including seven crypto assets (Bitcoin (BTC), Bitcoin Cash (BCH), Ether (ETH), Litecoin (LTC), Tether (USDT), TrueUSD (TUSD), USD Coin (USDC)) and monitors transactions of 800 exchanges (drafting time: June 2026).
- Data sources for NFTs:
  - Marketplaces and blockchains provide near-complete transaction histories; however, buyer anonymity and lack of residency/sectoral metadata impede direct macroeconomic compilation.
  - Tax filings, custodial records, and targeted surveys (creators, households) are necessary complements.
- DGI-3 Recommendation 11 templates (Rec. 11 TT) — scope and templates:
  - Template for CBDC issuance and holdings.
  - Template for Stablecoin Issuers: requires holder data by institutional sectors and by country for nonresident holders.
  - Template for Holdings and Flows of Crypto Assets: two parts — broad categories flows/positions and crypto-by-crypto holdings for seven assets (Bitcoins, Ether, Tether, BNB, SOL, USDC, XRP).
  - Short-term compilation: use custodial financial reports and Annex 3.5 guidance; flows estimated per Annex 3.3, accounting for loss/theft/seizures.
- Survey and reporting recommendations:
  - Annex 3.6 provides a Model Survey Questionnaire for Crypto Exchanges to collect positions (opening, net transactions, revaluations, other changes, closing) and breakdowns of assets under custody (including NFTs) with resident/nonresident splits and revenue/expenditure detail.
  - Annex 3.2 and Annex 3.4 provide model questionnaires for mining companies and stablecoin issuers respectively; Annex 3.8 and Annex 3.9 include enterprise/household and NFT-creator modules.

### Country practices and illustrative datasets (selected findings and numbers)
- Brazil (Central Bank of Brazil — BCB) ITRS implementation and 2025 aggregates (based on ITRS data):
  - Total purchases: US$ 18.3 billion
  - Total sales: US$ 0.8 billion
  - Purchases of crypto assets with corresponding liability: US$ 17.3 billion
  - Sales of crypto assets with corresponding liability: US$ 0.4 billion
  - Purchases with corresponding liability represented 97 percent of the net purchases in 2025.
  - BCB extraction algorithm: in last quarter of 2025, currency names recorded in approximately 80 percent of crypto asset purchase records and 67 percent of sale records; target to reach 100% coverage.
  - New reporting procedures for resident Virtual Asset Service Providers (VASPs) expected as of June 2026; VASPs required to report transaction-by-transaction cross-border details and, later, positions by type.
- Estonia (Eesti Pank) 2024 observations:
  - Number of service providers’ customers varied between 1.4 and 1.6 million.
  - Out of 45–50 service providers, only 3–5 companies had more than 100,000 customers.
  - Customer crypto assets exceed 2 billion euros, of which around 98 percent belonged to nonresident customers.
  - Over 90 percent market share formed by four companies; over 85 percent customers are natural persons; 10–15 percent are legal persons.
  - Asset shares: Bitcoin one-third of customers’ total assets; Ether 15 percent; Tether around four percent; other crypto assets 35–40 percent.
- Germany (Bundesbank) reporting framework:
  - Direct reporting mechanism with €50,000 threshold; since 2025 includes crypto asset transactions.
  - Four new reporting codes for crypto assets:
    - I. Crypto assets without corresponding liability — nonfinancial assets.
    - II. Foreign crypto assets with corresponding liabilities — financial assets.
    - III. Domestic crypto assets with corresponding liabilities — financial assets.
    - IV. Nonfungible tokens (NFTs) — nonfinancial assets.
  - Decision tree for classification emphasizes identifying counterpart geography and asset type; default assumptions when counterpart or issuer unknown lead to use of unallocated country code W19 or foreign asset classification.
  - Medium-term plans: implement coding to identify specific crypto asset within transactions (Digital Token Identifier) and consider obtaining data directly from major market players.
- Georgia (Geostat and National Bank of Georgia):
  - Georgia became a prominent mining hub due to low hydroelectric electricity costs and Free Industrial Zones; surge between 2017 and 2021 transformed Georgia from net exporter to net importer of electricity.
  - Legislative reforms effective January 1, 2023; as of October 2025 there are 28 registered VASPs.
  - VASPs obliged to comply with Travel Rule by end 2026; National Bank of Georgia developing remote supervisory reporting framework for VASPs.
  - Data collection challenged by uncooperative miners; electricity consumption used as imperfect proxy for mining scale.
- Illustrative corporate and custodial holdings (as reported in Annex 3.7 and Annex 3.5):
  - Example company reported holdings as of December 31, 2024:
    - MicroStrategy digital asset holdings (primarily Bitcoins) valued at $23.9 billion.
    - Tesla digital asset holdings (primarily Bitcoins) valued at $1.076 billion.
  - Estimated Coinbase Total Assets Under Custody as of December 31, 2024:
    - Total AUC: 404.0 ($bn)
      - Bitcoin: 235.4 ($bn) — US Residents (83%) 195.4 ($bn); Non-US (17%) 40.0 ($bn)
      - Ether: 54.2 ($bn) — US Residents (83%) 45.0 ($bn); Non-US (17%) 9.2 ($bn)
      - Solana: 21.3 ($bn) — US Residents (83%) 17.7 ($bn); Non-US (17%) 3.6 ($bn)
      - Other: 87.1 ($bn) — US Residents (83%) 72.3 ($bn); Non-US (17%) 14.8 ($bn)
      - USDC: 6.1 ($bn) — US Residents (83%) 5.1 ($bn); Non-US (17%) 1.0 ($bn)
    - Total split: 404.0 ($bn) — US Residents 335.3 ($bn); Non-US 68.7 ($bn)
    - Estimated sectoral allocation proxy for resident clients (Total AUC for resident clients: 335.3 ($bn)):
      - Financial Corporations (80%) and Households (20%):
        - Total for Financial Corporations: 268.4 ($bn); Households: 67.1 ($bn)
- Selected seizure events (Annex 3.1) — values as reported:
  - USA — Bitcoin — $3.6 billion — Bitfinex Hack (2022)
  - USA — Bitcoin — $3.36 billion — Silk Road (2021)
  - United Kingdom — Bitcoin — $3 billion — Money laundering operation
  - Germany — Bitcoin — $2.55 billion — Piracy website and laundering after Bybit hack
  - USA — Crypto assets — $225.3 million — Crypto Investment Scams (2025)
  - Spain — Crypto assets — EUR 27 million — Transnational criminal organization
  - Albania — Crypto assets — $10 million — Organized crime group
  - Netherlands — Crypto assets — $7.6 million — Illicit exchanges shut down
  - Australia — Bitcoin — $2.6 million — Stolen from French crypto exchange
  - Israel — Crypto assets — $1.7 million — Linked to Hezbollah and Iran's Quds Force
  - Serbia — Crypto wallets — $1 million — Fake crypto investment schemes

### Compilation recommendations, international cooperation, and data-sharing models
- Short-term compiler actions:
  - Leverage regulatory reporting, tax data, ITRS (where relevant), corporate financial statements, commercial on-chain analytics, and targeted surveys to assemble positions and flows.
  - Apply proxy methods and document assumptions when residency and sectoral breakdowns are missing (examples: custodial revenue as geographic proxy; trading volume splits as sectoral proxies).
  - Derive transactions and revaluations from position changes using market indices and observable prices (Annex 3.3 Tether and BTC illustrative steps).
- Medium-term actions:
  - Develop standardized reporting templates for mining companies, stablecoin issuers, and custodians (Annex 3.2; 3.4; 3.6).
  - Collaborate with regulators and supervisory authorities to improve disclosures (assets by instrument, currency, maturity, residency).
  - Enhance household and enterprise surveys to capture holdings and flows (Annex 3.8; 3.9).
- International coordination:
  - Recognize concentration of service providers across a few economies and the need for cross-border data exchange.
  - Create frameworks for standardized reporting and data sharing modeled on CPIS/CDIS to capture residents’ holdings with nonresident custodians.
  - DGI-3 Recommendation 11 templates provide an international standard for CBDC, stablecoin issuers, and holdings/flows of crypto assets; coordinated global adoption and data-sharing mechanisms are recommended.
- Data quality caveats:
  - Pseudonymity on public blockchains, off-chain activity (CEX custodial movements), and rapid market innovation complicate residency attribution and full coverage.
  - Commercial data providers use heuristics and proprietary methods; compilers must assess fitness for BPM7/2025 SNA purposes and triangulate multiple sources.

*Source: IMF Compilation Guide on Crypto Assets — Preface and Acknowledgements; Chapter I overview and Acronyms; Box 1.3; Boxes 2.3, 2.8, 2.10; Chapter 3 (Data Sources and Compilation Methods) and annexes; selected country experiences in Chapter 4 (Brazil, Estonia, Germany, Georgia).*

### PREFACE AND ACKNOWLEDGEMENTS [TO BE ADDED] _______________________________________ 5

### PREFACE AND ACKNOWLEDGEMENTS [TO BE ADDED]

### Purpose and scope
- The Guide states its primary purpose is to offer practical guidance on source data and compilation methods for measuring crypto asset-related transactions and positions in macroeconomic statistics, consistent with the System of National Accounts 2025 (2025 SNA) and the Integrated Balance of Payments and International Investment Position Manual, seventh edition (BPM7).
- The Guide supports implementation of the 2025 SNA/BPM7 treatment of crypto assets and the G20 Data Gaps Initiative (DGI-3) Recommendation 11 on digital money (including central bank digital currencies, stablecoins, and other crypto assets used as means of payment).
- National compilers are encouraged to adapt the Guide’s recommended data sources and methods to national circumstances to produce consistent and comparable estimates across economies.
- Chapter organization in the Guide:
  - Chapter I: Overview of crypto assets ecosystem (Purpose, Background, Current State of Play, Regulatory/Legal Frameworks, Organization of Guide).
  - Chapter 2: Methodological framework for recording crypto assets (typologies, issuance, other changes, exchanges/platforms, lending/borrowing, NFTs).
  - Chapter 3: Data sources and compilation methods (administrative, financial reports, surveys, commercial sources, NFTs, DGI-3 templates, data sharing).
  - Chapter 4: Country experiences and international initiatives; glossary and annexes provide examples, survey templates, and numerical recording examples.

### Key challenges and institutional context
- The Guide notes significant challenges in compiling crypto assets data arising from inherent complexities and limited access to suitable data sources; these challenges were recognized during methodological discussions and consultations.
- The Advisory Expert Group on National Accounts (AEG) and the IMF Committee on Balance of Payments Statistics (BOPCOM) recommended practical guidance to support implementation of 2025 SNA/BPM7 treatment of crypto assets.
- The IMF leads DGI-3 Recommendation 11; the Guide will support data collection following the Recommendation 11 task team templates (see Chapter 3, Section F).

### Current state of play — salient facts and figures
- The Guide reports the number of crypto assets has surged from a few hundred (including Bitcoin) to nearly 20,000.
- Market capitalization and volatility:
  - The total market capitalization of crypto assets stands at approximately $2  .5 trillion.
  - Bitcoin represents around 60 percent of that value.
  - The market reached a peak of $3 trillion in November 2021 and subsequently dropped below $1 trillion.
- Systemic risk considerations:
  - While crypto assets are not currently a major part of the global financial system, they are emerging as a potential source of systemic risk in certain jurisdictions; higher adoption and weak regulatory frameworks—especially in emerging and low-income economies—can elevate macro-financial and systemic risks through currency substitution, capital outflows, and interlinkages with the financial system.
- Macro-financial uses and risks:
  - Crypto assets may be used for cross-border payments, including remittances, sometimes offering faster and cheaper transactions.
  - Risks include money laundering, tax avoidance, and potential impacts on monetary policy transmission if adopted as means of payment.
  - Regulatory uncertainty remains a significant challenge limiting broader integration into the global financial system.

### Crypto assets ecosystem — definitions and components
- The Guide defines crypto assets as digital representations of value that use cryptography and distributed ledger technology (DLT) to enable peer-to-peer transactions without a trusted intermediary.
- Distinctions:
  - Fungible crypto assets: divisible and not unique (examples: Bitcoin, Ether, stablecoins such as USD Coin and Tether).
  - Nonfungible crypto assets (NFTs): unique and nondivisible (examples: Bored Ape Yacht Club, CryptoPunks, NBA Top Shot).
  - Native crypto assets: originate on their own blockchain and do not represent external assets (examples: Bitcoin and Ether).
  - Non-native crypto assets: represent underlying financial or real-world assets via tokenization (examples: tokenized equities, bonds, real estate; crypto securities).
- Core elements of the ecosystem (not mutually exclusive in operations):
  - Distributed ledger technology (e.g., blockchain)
  - Miners and validators
  - Issuers of stablecoins and NFTs
  - Crypto exchanges, custodians, wallet providers
  - Decentralized finance (DeFi)
  - Crypto payment processors

### Distributed ledger technology and blockchain characteristics
- DLT records data in multiple places simultaneously so participants have consistent access to the same data; in principle, DLT is designed to operate without a central authority acting as a trusted intermediary.
- Blockchain is a type of DLT organizing data in a chain of verified and validated blocks; blockchains can be public (usually permissionless and decentralized) or private (permissioned and centralized).
- Public blockchains feature pseudonymity: transactions are public and traceable, but parties are identified only by public addresses, complicating direct identification of holders’ residency and institutional sector for macroeconomic statistics.
- Crypto key concepts explained:
  - Private key: secret cryptographic key used to sign transactions and prove ownership; a Bitcoin private key is typically 64 characters long; loss of the private key means permanent loss of access to associated crypto assets.
  - Public key: cryptographic key derived from a private key; used to encrypt or verify signatures; cannot be used to derive the private key.
  - Public address: unique code derived from a public key used to receive crypto assets; functionally comparable to an account number on the blockchain.

### Boxes and annex material referenced (topics summarized)
- Box topics included: centralized vs decentralized ledger systems; public keys/public addresses/private keys; data available from blockchains; consensus mechanisms in public and private blockchains; crypto exchanges and on-chain/off-chain transactions.
- Annexes and numerical examples provide practical recording illustrations and survey templates (examples include numerical examples on mining/validation services, staking, pooled mining, revaluation, exchange/platform sectorization, crypto lending, NFT recording; model survey questionnaires for mining companies, stablecoin issuers, exchanges; estimated crypto holdings as of December 31, 2024; and a summary of 2025 SNA/BPM7 recommendations for crypto assets).

*Source: IMF Compilation Guide on Crypto Assets — Preface and Acknowledgements; Chapter I overview and Acronyms sections.*

### Box 1.3. Data from Blockchains

### Box 1.3. Data from Blockchains

### On-chain data availability and pseudonymity
- Data for on-chain crypto activity are available directly from blockchains, such as Bitcoin blockchain.
- Public blockchains provide data on balances and transaction histories for public addresses, but there is no direct way to link a blockchain address with the real identity of holders; this feature is referred to as pseudonymity.

### Types of data available from blockchains (Bitcoin example)
- Public addresses activity:
  - Current balance associated with a specific address,
  - A record for all incoming and outgoing transactions (with the addresses of sender and receiver)
- Individual transaction details:
  - Transaction unique identifier (transaction ID),
  - Sender’s and receiver’s pseudonymous addresses,
  - Crypto assets amount transferred and transaction fees,
  - Timestamp (date and time when the transaction was processed)
- Block data (transactions are validated and added to the block which is the element of blockchain):
  - Block hash (unique identifier for each block),
  - Block height (block number in order from the beginning of the network i.e., position of the block in the blockchain),
  - Miner information i.e., address of the miner who mined the block,
  - Number and list of transactions included in the block,
  - Timestamp i.e., date and time when the block was mined,
  - Size of the block in bytes.
  - Nonce (number used once in the mining process to obtain a valid block hash)
- Network data:
  - Total number of network nodes (computers that participate in the network and maintain the copy of the blockchain),
  - Total number of transactions processed during a period of time,
  - Hash rate - a measure of the computational power being used by miners to process and validate transactions by adding a block to the blockchain.

### Consensus mechanisms and network security
- A consensus mechanism is a protocol that ensures all participants in a decentralized network agree on the validity of transactions and the current state of the blockchain; it prevents double spending and maintains security and integrity.
- The most important consensus mechanisms in public blockchains are proof-of-work (PoW) and proof-of-stake (PoS).
- Proof-of-Work (PoW):
  - Miners solve complex mathematical puzzles to validate transactions and add new blocks.
  - The first miner to solve the puzzle can add the next block and receives crypto assets as a reward.
  - PoW secures the network by making alterations computationally difficult and resource-intensive.
  - Known for high energy consumption.
  - Examples: Bitcoin and Ether (before its transition to PoS).
- Proof-of-Stake (PoS):
  - Validators are chosen based on the number of coins they hold and lock up as "stake".
  - Validators are selected to create new blocks and validate transactions; they risk losing staked coins if they act maliciously.
  - PoS is more energy efficient compared with PoW.
  - Examples: Ether (after its transition to PoS) and Cardano.
- Private blockchains:
  - Operate within a controlled environment with verified and trusted participants.
  - Consensus mechanisms prioritize efficiency and speed, often revealing node identity and staking reputation rather than relying on PoS or PoW.
  - Proof of Authority (PoA) and Practical Byzantine Fault Tolerance (PBFT) are used in private blockchains.
  - PBFT requires a minimum of two-thirds of the nodes to agree and is suitable for small networks.
  - Examples: VeChain and Xodex use PoA; Binance Smart Chain and Ripple utilize PBFT.

### Miners, validators, and incentives
- In PoW-based blockchains (e.g., Bitcoin), miners use computational power to solve puzzles; once solved, a miner earns the right to validate pending transactions, propose the next block, and receive explicit transaction fees and a reward in the form of newly minted crypto assets.
- In PoS blockchains (e.g., Ethereum post-transition), validators are selected based on staked tokens and receive staking rewards and transaction fees for maintaining the network.
- Examples of large Bitcoin mining companies are reported at the time of drafting the Guide.

### Issuers of stablecoins
- Stablecoin issuers are entities other than monetary authorities or government that create, issue, and manage crypto assets designed to act as a general medium of exchange.
- Stablecoins may be issued by deposit-taking corporations (e.g., commercial banks such as JPMorgan’s JPM Coin) or by nonbank issuers such as Tether Limited and Circle.
- Stablecoin designs:
  - Backed by assets of the issuer (reserves), usually held in liquid financial instruments such as cash, deposits, and short-term securities; reserves may sometimes include nonfinancial assets.
  - Algorithmic stablecoins control supply via algorithms.
  - Hybrid approaches combine backing and algorithmic mechanisms.
- Use cases: trading, remittances and other transfers, serving as a general medium of exchange.
- Examples:
  - Tether (USDT): pegged to USD, the largest global stablecoin by market capitalization, issued by Tether Limited,
  - USD Coin (USDC): pegged to USD, issued by Circle and Coinbase,
  - Binance USD (BUSD): issued by Binance in partnership with Paxos, pegged to USD.

### Issuers of Nonfungible Tokens (NFTs) and NFT mechanics
- NFT issuers include individuals, organizations, or platforms creating and distributing NFTs representing digital or physical assets (art, music, collectibles, virtual real estate, in-game items).
- Examples of issuers and products:
  - Yuga Labs: issuer of Bored Ape Yacht Club (collection of hand-drawn ape NFTs granting owners exclusive community benefits),
  - Nike: issues virtual sneakers and digital wearables as NFTs for metaverse applications,
  - Decentraland: issues virtual land NFTs allowing buying, selling, and building on parcels of digital real estate.
- NFT attributes and lifecycle:
  - Each NFT requires a unique identifier referencing asset information (name, description, creator).
  - Creation (minting) is usually via execution of a smart contract, which generates a unique token and records it on the blockchain (e.g., Ethereum).
  - Ownership history (transactions of buys, sells, transfers) is recorded on the blockchain and is publicly accessible, allowing verification of current and past owners.

### Crypto exchanges, custodians, and wallet providers
- Crypto exchanges are digital platforms where users can buy and sell crypto assets, including NFTs; they may offer custody, transactional and wallet services, crypto lending and borrowing.
- Custody refers to secure storage and management of private keys; custodians protect private keys using hot and cold storage solutions.
- Custodial service providers may include financial corporations and banks that do not operate as exchanges (examples cited in the Guide).
- Crypto wallets:
  - Hot wallets: connected to the internet, quick access, more vulnerable to hacks.
  - Cold wallets: offline custody of private keys, enhanced security for long-term storage.
  - Non-custodial (self-custody) wallets: users retain full control over private keys. Examples include Ledger, Trezor, and MetaMask.
- Overlap exists among exchanges, custodians, and wallet providers; some entities offer multiple services (example cited in the Guide).

### CEX vs DEX, and on-chain vs off-chain transactions
- Two main types of exchanges:
  - Centralized exchanges (CEX): mostly custodial, built around a centralized order book, often subject to regulatory oversight and KYC requirements; examples cited in the Guide.
  - Decentralized exchanges (DEX): non-custodial, leverage smart contracts to facilitate direct peer-to-peer trading, with private keys held by users and transactions recorded on-chain.
- Transaction categories:
  - On-chain transactions:
    - Occur directly on the blockchain and are recorded on the distributed ledger,
    - Immutable and transparent (limited by public blockchain pseudonymity),
    - Potentially slower and characterized by higher operational cost (includes transaction fees paid to miners or validators),
    - Commonly associated with DEXs.
  - Off-chain transactions:
    - Occur outside the blockchain and are not immediately recorded on the distributed ledger,
    - Faster processing times and lower fees,
    - Not publicly visible, offering greater privacy,
    - Examples include transactions within CEXs and certain payment channels.
- Table 1.1 comparison (summarized in prose):
  - Feature: Custody — CEX: Custodial; DEX: Non-Custodial
  - Feature: Transaction Type — CEX: Off-Chain and On-Chain; DEX: On-Chain
  - Feature: Transparency — CEX: Lower; DEX: Higher (but limited by public blockchains pseudonymity)
  - Feature: Regulation — CEX: Subject to regulation in some jurisdictions, in particular often KYC requirements; DEX: Less regulated
  - Feature: User Experience — CEX: User-friendly; DEX: Varies

### Decentralized Finance (DeFi) and payment processors
- DeFi:
  - Encompasses services replicating traditional financial services on blockchains with smart contracts (e.g., Ethereum).
  - Business models focus on investment opportunities; direct impact on financing traditional real-world businesses or households is still limited.
  - Examples: lending/borrowing platforms like Aave and Compound; DEXs such as Uniswap and SushiSwap; insurance platforms like Nexus Mutual.
- Crypto payment processors:
  - Enable businesses to accept crypto asset payments, offering benefits like lower transaction fees, faster cross-border payments, enhanced security.
  - Examples of business models include BitPay and CoinPayments.

### Regulatory and legal frameworks overview
- Regulatory approaches vary widely across jurisdictions: some embrace crypto assets with limited regulation, others impose restrictions, and some ban private crypto assets trading.
- Economies working on and implementing comprehensive regulatory approaches include Japan, Singapore, South Korea, and Switzerland.
- The European Union legal act Markets in Crypto Assets Regulation (MiCAR) aims to create a harmonized framework across EU member states with specific rules for stablecoins.
- In the USA, a federal law enacted in July 2025 — Guiding and Establishing National Innovation for U.S. Stablecoin Act (the GENIUS Act) — provides a comprehensive regulatory framework for payment stablecoins.
- China has taken a restrictive approach and banned private crypto assets trading.
- A survey conducted as part of Recommendation 11 of DGI-3 in February-March 2024 covered questions on crypto assets and shed light on differences in regulatory frameworks across G20 and FSB economies participating in DGI-3.

*Source: The Guide (International Monetary Fund).*

### 1.27 The analysis of the results shows that regulations on stablecoins and other crypto assets exist (or

### crypto-assets-cg - 1.27 The analysis of the results shows that regulations on stablecoins and other crypto assets exist (or

### Regulatory landscape and obligations
- The Guide reports that regulations on stablecoins and other crypto assets exist (or are under implementation) in around 30 percent of G20/FSB participating economies.
- Several economies reported regulations that ban or significantly limit the use of private crypto assets, especially crypto assets without a corresponding liability designed to act as a medium of exchange:
  - In those economies, private crypto assets are either forbidden or cannot be used as means of payments.
  - There are limitations on engagement of regulated and supervised intermediaries, mostly banks, in the crypto ecosystem.
- Several economies indicated regulatory obligations imposed on players in the crypto ecosystem, in particular CEXs, crypto digital wallet providers and other custodians.
  - Existing regulations frequently require AML/CFT procedures as a minimum requirement.
  - Comprehensive financial regulations for the crypto assets ecosystem are still to be developed and implemented in most economies.
  - More complex regulatory schemes cover AML/CFT and market integrity plus issues such as risk management, individual investor protection, and informational disclosure requirements.

### Organization and scope of the Guide
- Chapter structure summarized by the Guide:
  - Chapter 2: methodological framework for recording crypto assets in line with the 2025 SNA/BPM7, including classification and typology of fungible crypto assets; treatment of mining, validation, staking, and lending; recording of transactions, revaluations, and other changes in volume, including NFTs; institutional sectoring; output of crypto exchanges and platforms; cross-border recording issues.
  - Chapter 3: data sources and compilation methods—administrative and supervisory data, financial statements of crypto-related entities, commercial data sources, and NFT-specific data; role of international data sharing; annexes with practical compilation tools.
  - Chapter 4: country experiences and international initiatives; collaborative efforts led by international organizations.
  - The Guide concludes with a glossary of terms and definitions related to crypto assets.

### Typology and classification (2025 SNA/BPM7)
- Fungible crypto assets are classified into three broad categories:
  - crypto assets designed to act as a general medium of exchange;
  - crypto assets that only act as a medium of exchange within a platform or network;
  - security crypto assets.
- Recording guidance for fungible crypto assets with a corresponding liability:
  - Crypto assets designed to act as a general medium of exchange, with a corresponding liability:
    - issued by a monetary authority (e.g., central bank digital currencies (CBDCs) that qualify as crypto assets)—currency under “currency and deposits” (AF21).
    - not issued or authorized by monetary authorities or government (e.g., stablecoins with a claim on the issuer)—classified as a separate category under “currency and deposits” (AF22).
  - Crypto assets that only act as a medium of exchange within a platform or network (payment tokens):
    - with a corresponding liability—separate subcategory under “debt securities” (AF33—short-term and AF34—long-term).
  - Security crypto assets (which always have a counterpart liability):
    - Debt security crypto assets—“debt securities” (AF3).
    - Utility tokens—separate category under “debt securities” in the 2025 SNA; included under “debt securities” in the BPM7 (AF35).
    - Equity crypto assets—included under “equity” (AF51).
    - Derivative crypto assets—included under “financial derivatives” (AF71).
- Crypto assets without a corresponding liability designed to act as a general medium of exchange (CAWLM) and crypto assets without a corresponding liability designed to act as a medium of exchange within a platform or network (CAWLP) are classified as nonproduced nonfinancial assets and recorded as a separate category in the capital account of BPM7 and in the capital account and balance sheet of 2025 SNA (NF22).
- Equity crypto assets classification as listed shares (F511), unlisted shares (F512), or other equity (F519) should be based on the token’s own characteristics.

### Reserve asset treatment (BPM7 guidance)
- BPM7 recommends that crypto assets without corresponding liabilities designed to act as a general medium of exchange (e.g., Bitcoin, Ether) should not be included in reserve assets; they are nonproduced nonfinancial assets and holdings by a monetary authority should be recorded under a separate category under nonproduced nonfinancial assets.
- For crypto assets treated as financial assets (including stablecoins and security crypto assets) to qualify as reserve assets, they must satisfy the criteria in paragraphs 6.69–6.80, BPM7, including:
  - being readily available to and under the effective control of the monetary authorities in the most unconditional form to meet balance of payments financing needs;
  - representing actual, high-quality external claims on nonresidents;
  - being denominated and settled in convertible foreign currency.
- The Guide notes most stablecoins currently appear not to meet these requirements—particularly with respect to constituting high-quality external assets—but a case-by-case assessment is necessary.

### How fungible crypto assets enter circulation and recording conventions
- Creation/circulation mechanisms:
  - Miners validating transactions in crypto assets without a corresponding liability receive explicit validation fees and implicit fees (block rewards in terms of new crypto assets).
  - Non-mineable crypto assets without a corresponding liability may be released via explicit sale and/or as payment to validators in proof-of-stake or proof-of-authority systems.
- Treatment of mining/validation services:
  - Activities that create new crypto assets without a corresponding liability are regarded as production activities.
  - Services of miners/validators should be measured as the sum of explicit validation fees and implicit fees (newly released crypto assets).
  - Most producers of these services are likely classified in the nonfinancial corporations sector.
  - Explicit fees are paid by the party initiating the transaction; implicit fees are assumed to be collectively paid and consumed by existing holders of the crypto asset.
  - By convention, implicit fees are recorded as a cross-border transaction payable by the owners of existing crypto assets to the producer, with the new crypto assets diluting the value of existing crypto assets.
  - The convention may create global asymmetries, but it promotes consistent recording across countries given practical difficulties in identifying holders of existing crypto assets.

### Numerical example: recording of crypto mining/validation services (from the Guide)
- Scenario parameters:
  - Economy A mining company provides validation services for transactions originated in Economy B.
  - Economy B institutional units pay a transaction fee of 5 in crypto assets.
  - Economy A mining company receives 20 units of crypto assets as implicit fee.
  - The mining company incurs 18 crypto units on electricity and other intermediate inputs.
  - Market value of each crypto asset is USD 1, 000.
- Entries recorded in Economy A BOP (in USD):
  - Current account — Credit/Revenue:
    - Services — Computer services (explicit fee—from Economy B) 5,000
    - Services — Computer services (implicit fee—from rest of the world) 20,000
  - Capital account — Acquisition/disposal of non-produced nonfinancial assets:
    - Crypto assets without corresponding liabilities designed to act as a medium of exchange (from Economy B and rest of the world) 25,000 (5,000+20,000)
- Economy A National Accounts (in USD):
  - Production account — Output (explicit fee + implicit fee) 25,000
  - Intermediate consumption (electricity and other inputs) 18,000
  - Gross value added 7,000
  - Capital account — Change in assets:
    - Acquisition/disposal of non-produced nonfinancial assets — Crypto assets without corresponding liabilities designed to act as a medium of exchange 25,000 (5,000+20,000)
  - Balance sheet — Assets:
    - Non-produced nonfinancial assets — Crypto assets without corresponding liabilities designed to act as a medium of exchange 25,000
- The Guide explains that the implicit fee of 20 units ideally should be distributed to economies holding the crypto assets based on allocations, but due to practical limitations the counterpart is conventionally recorded as the whole crypto community considered nonresident.

### Staking, pooled and cloud mining (treatment)
- Crypto staking:
  - Staking: holders allocate crypto assets in a designated wallet to participate in a PoS network; validators are chosen based on staked amounts.
  - Participation modes:
    - Active staking (direct via blockchain): accrued revenue for stakers (block rewards (implicit fee) and explicit transaction fees) should be treated as the output of validation services and recorded as computer services in the balance of payments.
    - Passive staking (through CEX/DEX): staking revenue recorded on a gross basis; fees imposed by platforms are recorded as intermediate consumption for provision of staking-as-a-service and classified as computer services for practical reasons.
- The implicit fee in staking is assumed payable by owners of existing crypto assets to producers of the services, with new crypto assets diluting existing units (parallel to PoW treatment).

*Source: The Guide.*

### Box 2.3. Numerical Example on the Recording of Crypto Staking

### Box 2.3. Numerical Example on the Recording of Crypto Staking

### Scenario description
- Institutional unit A from Economy A places 1000 units of a stablecoin called Stabex (a stablecoin with a corresponding liability issued by Economy C and one Stabex=1 USD) with a CEX called Cryptostake in Economy B for (passive) staking.
- Cryptostake performs validation of Stabex transactions (i.e., undertakes passive staking on behalf of clients from Economy A) and earns staking revenue of 150 units in Stabex.
- Cryptostake retains 30 units as fee and pays the remaining 120 units to Institutional Unit A.
- Cryptostake pays 5 units of Stabex on electricity and other intermediate inputs to domestic institutional units, in relation to staking.
- All payments are made in USD.

### Balance of payments entries — Economy A (passive staker, Institutional Unit A)
- Current account balance: 120
  - Credit/Revenue: Services — Computer services (staking revenue): 150
  - Debit/Expenditure: Services — Computer services (fee to Economy B): 30
- Financial account (Net): 120
  - NAFA* NIL*
  - Other investment — Currency and deposits (vis-à-vis Economy C): 120

### Balance of payments entries — Economy B (validator, Cryptostake)
- Current account balance: 30
  - Credit/Revenue: Services — Computer services (validation): 30
  - Debit/Expenditure: (no itemized debit shown beyond intermediate inputs in national accounts)
- Financial account (Net): 30
  - NAFA* NIL*
  - Other investment — Currency and deposits: 30

### National accounts — Economy A (passive staker)
- Production account:
  - Expenditure: Intermediate consumption (fee to Economy B for services): 30
  - Revenue: Output (passive staking rewards): 150
  - Gross value added: 120
- Financial account:
  - 120
  - NAFA* NIL*
  - Currency and deposits: 120

### National accounts — Economy B (validator, Cryptostake)
- Production account:
  - Revenue: Output (fee from Economy A): 30
  - Expenditure: Intermediate consumption: 5
  - Gross value added: 25
- Financial account:
  - 25
  - NAFA* NIL*
  - Currency and deposits: 25

### Key numeric facts and flows (preserved exactly as reported)
- Stabex placed: 1000 units (1 Stabex = 1 USD)
- Staking revenue earned by Cryptostake: 150 units
- Fee retained by Cryptostake: 30 units
- Payment to Institutional Unit A: 120 units
- Intermediate inputs paid by Cryptostake domestically: 5 units
- Economy A BOP current account balance (passive staker perspective): 120
- Economy B BOP current account balance (validator perspective): 30
- Economy A national accounts: Output 150; Intermediate consumption 30; Gross value added 120; Financial account 120
- Economy B national accounts: Output 30; Intermediate consumption 5; Gross value added 25; Financial account 25

*International Monetary Fund (IMF).*

### 2.28 Crypto lending is a process where institutional units lend their crypto assets to other institutional

### Crypto lending and NFTs (Crypto Assets Guidance)

### Crypto lending: definition and channels
- Crypto lending is a process where institutional units lend their crypto assets to other institutional units for a specified period in exchange for an agreed payment/revenue in crypto assets or in fiat currencies.
- The borrower can use the borrowed assets for lending, trading, investing, and liquidity management.
- Crypto lending applies to both:
  - crypto assets with a corresponding liability (e.g., some stablecoins), and
  - crypto assets without a corresponding liability (e.g., Bitcoins, Ether).
- Mostly channeled through crypto exchanges/platforms (centralized and decentralized) that:
  - take custody of deposited assets,
  - set payment rates,
  - handle collateral, and
  - manage the lending/borrowing process.
- Centralized platforms (e.g., Nexo) are generally described as off-chain lending; decentralized platforms (e.g., Aave) are generally described as on-chain lending.

### Recording and statistical treatment of crypto lending
- Crypto assets without a corresponding liability designed to act as a general medium of exchange:
  - Fees payable to the owners used for lending should be recorded as interest (by convention, on the lines of fees for securities lending and gold loans).
  - Institutional units that lend these assets remain the economic owner (no transaction in those crypto assets is recorded in the capital account and no changes are made to the balance sheets of the parties).
  - Institutional units that borrow these assets become legal owners and can on-lend them to others.
- Lending of crypto assets with a corresponding liability (security crypto assets):
  - Treated as financial instruments; treatment for traditional securities lending applies.
  - No transaction is recorded in the financial account if the lender remains the economic owner.
- Lending of stablecoins (crypto assets with a corresponding liability designed to act as a general medium of exchange—not issued or authorized by monetary authorities or government) and CBDCs on the blockchain:
  - Should be recorded along the lines of cash lending.
  - Associated lending revenue/payment is recorded as interest.

### Numerical example on recording of crypto lending (Box 2.8)
- Institutional unit A (Economy A) lends 100 Bitcoins to Institutional unit B (Economy B) and receives the principal amount along with 5 Bitcoins as revenue after one year.
- Institutional unit A lends 10,000 units of Tether to a unit in Economy C and receives 500 Tether as interest payment for one year.
- Prices: one Bitcoin = 1,000 USD; one Tether = 1 USD.

Entries recorded for Economy A — first transaction (Bitcoin) (USD):
- Current account — Credit/Revenue: Earned Income — Interest: 5,000
- Capital account — Debit/Expenditure: Acquisitions/disposals of nonproduced nonfinancial assets — Crypto assets without a corresponding liability designed to act as a medium of exchange: 5,000

National accounts for Economy A — first transaction (USD):
- Allocation of Earned Income Account — Interest: 5,000
- Capital account — Change in assets and net worth — Acquisitions/disposals of nonproduced nonfinancial assets — Crypto assets without a corresponding liability designed to act as a medium of exchange: 5,000

Entries recorded for Economy A — second transaction (Tether) (USD):
- Current account — Credit/Revenue: Earned Income — Interest: 500
- Financial account:
  - NAFA* NIL*
  - Other investment — Currency and deposits (Tether) -10,000
  - Currency and deposits (Tether) 500
  - Loans (Tether) 10,000

National accounts for Economy A — second transaction:
- Allocation of Earned Income Account — Interest: 500
- Financial account — Change in assets and net worth:
  - Currency and deposits (Tether) -9,500 (-10,000+500)
  - Loans (Tether) 10,000

- *NAFA – Net acquisition of financial assets, NIL – Net incurrence of liabilities

### Nonfungible tokens (NFTs): definition and distinguishing features
- NFTs are digital records hosted on a blockchain associated with a digital or physical asset or product but distinct from that asset or product.
- NFTs certify rights to use and benefit from the asset and may certify authenticity.
- Each NFT has a distinct identifier stored on the blockchain with metadata (creator's information, creation date, and other attributes).
- NFTs are non-divisible: payments for NFTs usually must be made in the fungible crypto asset native to the blockchain on which the NFT is hosted.

### Classification of NFTs (2025 SNA/BPM7)
- All NFT issuers are classified as nonfinancial corporations.
- Three types of NFTs:
  - First type: convey no ownership rights and only allow for personal use of a specified asset or product.
  - Second type: convey limited ownership rights beyond personal use for a specified asset or product.
  - Third type: convey full ownership rights for a specified asset or product.

### Statistical treatment by NFT type
- First type:
  - Record as final consumption expenditure in national accounts.
  - Cross-border transactions recorded under the relevant service category (computer services, audiovisual and related services, or information services) depending on the content.
  - In rare cases, may transform into a valuable later (similar to objects of art).
- Second type:
  - Treated as nonproduced nonfinancial assets (included in asset class “contracts, leases and licenses”) and recorded in the capital account if they confer economic benefits that the holder can realize.
- Third type:
  - An NFT conveying full ownership is a digital recording of ownership similar to a property title, not a separate asset.
  - The underlying asset should already be recorded in the integrated framework of national accounts; purchasing such an NFT is a way of purchasing the underlying asset.
  - Cross-border transactions should be recorded as computer services, audiovisual and related services, or information services if the underlying asset is digital.

### Examples and numerical example for NFTs (Box 2.10)
- Examples by type (Table 2.1):
  - First type: NBA Top Shot “Moments” — Short video clips and stats of NBA players — Personal use and display rights.
  - Second type: CryptoPunks — Pixel-art digital characters — Ownership of the NFT and display rights.
  - Third type: Florida, USA house sold via NFT (Feb 2022) — Real estate property — Ownership via transfer of the NFT.
- Numerical example (Box 2.10):
  - A resident of Economy A sells CryptoPunk #4567 to a resident of Economy B via the official CryptoPunks marketplace operated by Yuga Labs in Economy C.
  - NFT valued at USD 300,000. Seller pays a fee of 0.5% to marketplace. Payment made in Ether. No direct fee charged to buyers.
  - BOP entries for Economy A (USD):
    - Current account: -1,500 — Services — Nonfinancial intermediation services (payments to marketplace in Economy C) — 0.5%*300,000 =1,500
    - Capital account: 1,500 — Acquisitions/disposals of nonproduced nonfinancial assets — Contracts, leases, and licenses (NFTs-second type) vis-à-vis Economy B: 300,000
    - Crypto assets without a corresponding liability designed to act as a medium of exchange (Ether) vis-à-vis Economies B (300,000) and C (1,500): 1,500 300,000
  - Outcome:
    - Seller from Economy A receives $298,500 after paying the fee of $1,500 to the platform (Yuga Labs) in Economy C.
    - The fee ($1,500) charged by Yuga Labs is treated as import of nonfinancial intermediation services from Economy C.
  - National accounts of Economy A:
    - Disposable Income Account — Final consumption expenditure (payment for nonfinancial intermediation services to Economy C): 1,500
    - Capital account — Change in assets and net worth:
      - Acquisitions/disposals of nonproduced nonfinancial assets — Contracts, leases, and licenses (NFTs-second type) vis-à-vis Economy B: -300,000
      - Crypto assets without a corresponding liability designed to act as a medium of exchange (Ether) vis-à-vis Economies B (300,000) and C (1,500): 300,000-1500=298,500

### Annex summary points (selected)
- Crypto assets with a corresponding liability designed to act as a general medium of exchange that are issued by a monetary authority (AF21) — recorded as currency under “currency and deposits.”
- Crypto assets with a corresponding liability designed to act as a general medium of exchange that are not issued or authorized by monetary authorities or government (AF22) — separate category under “currency and deposits.”
- Crypto assets without a corresponding liability designed to act as a medium of exchange (AN22) — separate category under “nonproduced nonfinancial assets”—capital account/balance sheet (examples: Bitcoin, Ether).
- Crypto exchanges/platforms (including those offering lending and borrowing facilities) are classified as financial corporations (financial auxiliaries) or nonfinancial corporations (nonfinancial intermediaries) depending on dominant activity.
- Outputs of crypto mining, staking, pool mining, and cloud mining are recorded as computer services; output measures include explicit and implicit fees and block rewards as applicable.
- Crypto lending/borrowing summarized:
  - Lending of crypto assets without a corresponding liability: fee recorded as interest; crypto assets on lent remain on lender’s balance sheet.
  - Lending of crypto assets with a corresponding liability—security crypto assets: lending fee recorded as interest; crypto assets on lent remain on the lender’s balance sheet.
  - Lending of crypto assets with a corresponding liability—CBDCs on blockchain and stablecoins: recorded on the lines of cash lending; lending fee recorded as interest.

*Source: IMF — Crypto Assets Guidance (2025 SNA/BPM7 recommendations).*

### CHAPTER 3. DATA SOURCES AND COMPILATION METHODS

### CHAPTER 3. DATA SOURCES AND COMPILATION METHODS

### A. INTRODUCTION
- The chapter presents the range of data sources available for compiling crypto asset statistics, reflecting the diverse and complex nature of crypto-related activities.
- Key observation: although crypto assets are based on decentralized and often pseudonymous design, the crypto ecosystem exhibits meaningful functional and institutional centralization that can be leveraged for statistical purposes.
- Core activities concentrated in identifiable entities include: trading, custody, issuance, validation, and intermediation.
- Covered source types:
  - Administrative and supervisory data
  - Financial statements of mining companies and crypto exchanges
  - Alternative commercial and blockchain-based data
  - International data sharing possibilities
- Purpose: enable compilers to select suitable approaches to produce consistent and reliable statistics on crypto assets following the 2025 SNA/BPM7 methodological framework (see Chapter 2), taking into account institutional and legal frameworks for statistical reporting.
- Organization of chapter:
  - Section B: main administrative data sources (regulatory, tax, ITRS)
  - Section C: financial statements of mining companies, stablecoin issuers, crypto exchanges
  - Section D: alternative (commercial) data sources
  - Section E: data sources related to Nonfungible tokens (NFTs)
  - Section F: data sources and compilation approaches relevant to DGI-3 Recommendation 11 Task Team templates on stablecoins and other crypto assets
  - Section G: international data sharing to enhance consistency
  - Annexes: model survey questionnaires and mapping/compilation methods (Annex 3.2; 3.4; 3.6; 3.9; 3.8; 3.7; 3.5; 3.3)

### B. ADMINISTRATIVE DATA SOURCES — overview
- Administrative sources derive from government institutions' core functions: tax authorities, financial regulators, ITRS, and government records on crypto asset losses, thefts, and seizures.
- These sources can be used to compile external sector and national accounts statistics under BPM7/2025 SNA, supplement financial reports or surveys, and validate other sources.

#### 1. Crypto Asset Service Providers (CASPs)
- Definition: institutional units providing crypto-asset-related services (custody, exchange services, digital wallet operations, payment processing).
- Registration and oversight:
  - In several countries, CASPs register with central banks, financial regulators, securities regulators, or ministries of finance and provide operational details.
  - Regulatory agencies typically publish lists of CASPs covering domestic and local branches/subsidiaries of foreign platforms.
  - Example: in the United States, crypto exchanges register with FinCEN as money services businesses and comply with AML/CFT obligations, including suspicious activity reports.
- Sectoral implications:
  - Some CASPs fall under SEC or CFTC jurisdiction depending on activities, and could face additional disclosure obligations.
  - National statistical offices and central banks should coordinate with relevant agencies to compile CASP lists for potential inclusion in business registers.

#### 2. Regulatory Data on Crypto Assets
- Regulatory reporting by CASPs offers a practical starting point for compiling crypto asset data under BPM7 and 2025 SNA frameworks, especially if tailored for statistical reporting.
- Content of regulatory reporting:
  - Transaction-level data, AML reports, operational metrics, licensing and compliance documentation, and financial statements.
- Statistical relevance:
  - Datasets typically provide details on issuance, transfers, holdings, and counterparty relationships — useful for tracking cross-border flows, classifying crypto entities, and assessing financial interconnectedness and systemic risk.
- Limitations and challenges:
  - Regulatory filings often do not explicitly distinguish between crypto assets with and without corresponding liabilities.
  - May not provide sector and residency of crypto asset holders or details of entities involved in transactions.
  - Coverage gaps from unlicensed or decentralized entities.
  - Confidentiality restrictions may exclude specific details or transaction-level data.
  - Jurisdictional differences in reporting standards hinder harmonization and comparability.
  - Timeliness concerns: reports are often periodic and may lag market developments.
  - Inconsistencies in scope, definitions, and data availability across jurisdictions.
- Cross-reference: experience of Eesti Pank compiling CASP reports is discussed in Section B.2, Chapter 4.

#### 3. Tax Data on Crypto Assets
- Tax frameworks provide records on crypto-related transactions and income, useful especially for resident households and corporations.
- Uses of anonymized and aggregated tax data:
  - Assessing adoption and market activity:
    - Number of taxpayers reporting crypto activity as a proxy for adoption levels.
    - Reported transaction volumes as a baseline for market turnover.
    - Declared asset values for assessing net worth.
  - Estimating household income, consumption, and savings:
    - Reported income from mining, staking can refine household income/savings measurement.
  - Tracking cross-border financial flows:
    - Taxpayer disclosures of foreign-held crypto assets or transactions with nonresidents provide partial views of cross-border flows for financial account compilation in the BOP.
- Country examples and limitations:
  - U.S. tax data: captures flows of transactions and realized gains but does not provide wallet balances or total crypto holdings.
  - Canada: taxpayers must report investment income from crypto trades; no standalone requirement to disclose total holdings unless foreign asset rules are triggered.
  - India: income tax data provides disposals and related gains; no direct requirement to declare total domestic crypto balances.
- Constraints of tax data for macro statistics:
  - Tax data provide flows of income or realized gains/losses, not the asset transaction flows required for the financial account (for crypto assets classified as financial assets) or capital account (for crypto assets classified as nonfinancial assets).
  - May not capture peer-to-peer transfers, transactions on decentralized exchanges, and other non-custodial activities.
  - No information on total stock (holdings) of crypto assets.
  - Statistical concept of residence may differ from tax residency, affecting sectoral balance sheets and IIP compilation.
  - Frequency and timeliness: income tax reporting typically occurs annually.
  - Scope and methodology vary across countries due to unharmonized tax laws and divergent compliance levels.
- Potential improvements:
  - International initiatives like OECD’s Crypto-Asset Reporting Framework (CARF) and DAC8 in the EU may enhance usefulness of tax data for macroeconomic statistics in the future.

Box 3.1 — Leveraging OECD Crypto-Asset Reporting Framework (CARF) for Macroeconomic Statistics
- CARF features (reporting by Reporting Crypto-Asset Service Providers — RCASPs):
  - Acquisitions and disposals of crypto assets
  - Exchange transactions (crypto-to-crypto, crypto-to-fiat)
  - Transfers to and from wallets
  - Valuation and timestamps of each transaction
  - Identification of users and wallet addresses
- Challenges in using CARF data:
  - Data Privacy: access may be limited due to confidentiality rules, requiring anonymization or aggregation.
  - Coverage Gaps: non-compliant or decentralized platforms may not be captured.
  - Valuation Issues: volatile prices and inter-exchange price differences complicate consistent measurement.
  - Classification Complexity: distinguishing financial vs non-financial crypto assets can be challenging.
  - Timeliness: delays in reporting or international exchange can affect usability.
  - Standardization Across Jurisdictions: varying implementation timelines and interpretations may affect comparability.

#### 4. International Transactions Reporting System (ITRS)
- ITRS can capture cross-border transactions in crypto assets within BPM7 when crypto-related payments are processed through the domestic banking system.
- Data elements available when applicable: value, currency, date, and counterparty country of crypto asset transactions.
- Country practice: some countries (e.g., Brazil and El Salvador) have incorporated crypto-related categories and dedicated reporting codes into ITRS for purchases/sales of crypto assets and crypto-related services.
- Limitations of ITRS:
  - Records only transactions intermediated by domestic banks — excludes peer-to-peer trades and transactions conducted through overseas crypto exchange accounts.
  - Captures payment flows but not asset positions; requires complementary sources to measure residents’ crypto holdings.
  - Data quality depends on banks’ coding/reporting accuracy; may not distinguish asset types (with vs without corresponding liabilities).
  - Transactions through corporate accounts held abroad can bypass domestic banking system and fall outside ITRS coverage.
- Role: when integrated with complementary sources, ITRS contributes to BPM7-compliant measurement of cross-border crypto asset transactions.
- Cross-reference: Central Bank of Brazil experience compiling cross-border crypto transaction data from ITRS is in Section B.1, Chapter 4.

#### 5. Crypto Assets Loss, Theft, and Seizures
- Incidence: crypto asset seizures are increasingly common as law enforcement enhances capabilities; the United States has led many high-profile seizures, with other countries undertaking significant seizures.
- Data availability:
  - Annex 3.1 provides data on prominent seizures for selected countries gathered from news reports at drafting time.
  - Beyond news reports, seizure data are not systematically available for incorporation into national accounts and the balance of payments.
- Data providers and practices:
  - Hacking/theft/loss data are primarily compiled by private blockchain analytics firms (e.g., Chainalysis) using blockchain-tracing tools.
  - Law enforcement agencies (e.g., FBI) maintain internal records for criminal cases; academics and journalists aggregate public information on major events.
  - Data may be collected from law enforcement, financial intelligence units, customs, judicial institutions, and blockchain analytics firms.
  - Interagency data-sharing protocols are essential, especially for cross-border incidents, to ensure systematic capture and appropriate classification.
- Statistical treatment:
  - Loss, theft, hacking, and seizures are not considered transactions because they lack mutual consent and are treated as other changes in volume in balance of payments/national accounts (see Section E, Chapter 2 and Box 2.5).

### C. FINANCIAL REPORTS — overview
- Public financial reports of entities engaged in crypto activities (mining, stablecoin issuers, crypto exchanges, corporates holding crypto assets) provide data on flows and stocks and on income/expenses.
- These reports are typically prepared on a consolidated group basis but are useful for assessing domestic and cross-border crypto activities.

#### 1. Crypto Mining Companies
- Guidance is provided for compiling output, flows, and stocks of mining companies using publicly available financial statements of large MNEs engaged in mining.
- Financial statements can provide consolidated insights into mining and other activities, financial flows, and positions, including flows/stocks of crypto assets across subsidiaries in different countries.
- Topic indicator headings to be addressed (as in source): output and intermediate consumption (further details follow in the chapter beyond the provided excerpt).

*Source: CHAPTER 3. DATA SOURCES AND COMPILATION METHODS, IMF Guide on crypto-assets*

### 3.25 Crypto mining companies (e.g., Bitfury, Bitdeer, ClearSpark) typically earn revenues from two

### crypto-assets-cg - 3.25 Crypto mining companies (e.g., Bitfury, Bitdeer, ClearSpark) typically earn revenues from two

### Crypto mining companies — revenues, outputs, and secondary activities
- Revenue components:
  - Block rewards (newly mined crypto assets).
  - Transaction fees.
- Company activities and outputs:
  - Self-mining and provision of hosting and mining services (access to custom mining hardware and data center solutions).
  - Primary output recorded as mining/validation services and related activities.
  - Production of machinery and equipment or software solutions for mining should be recognized as secondary outputs.
- Data limitations for geographic and residency allocation:
  - In some cases, data on revenue components are available separately by subsidiary, allowing geographical distribution of revenues (mining activity).
  - Distinctions between resident and nonresident sources of transaction fees are often missing.
  - No systematic allocation of newly mined Bitcoins to specific countries is available.
  - Breakdowns of revenue from residents and nonresidents are generally not provided.
  - Additional revenues (hosting/mining for others) sometimes lack sufficient granularity at the subsidiary level to assign production to the economy accurately.

### Inputs and intermediate consumption for mining
- Main input:
  - Electricity is the main input and accounts for a substantial share of intermediate consumption.
- Expense reporting variability:
  - Some financial statements provide revenue and expenses by product lines with expense splits into electricity, depreciation, and other costs.
  - Others present expenses such as rent, wages, sales and general administration costs without separate electricity consumption data.
- Proxy and operational metrics:
  - Proxy approaches may be needed to derive electricity usage where direct data are missing.
  - Estimates based on data from similar mining companies can be used as proxies.
    - Example: data from Bitdeer shows that electricity consumption accounts for 25 percent of the total expenditure for each business line including mining.
  - Operational metrics such as hash rate, number of machines, and energy efficiency indicators can be useful but may not be readily available.
- Compilation caution:
  - Compilers must distinguish between expenses representing intermediate consumption (e.g., electricity) and those to exclude (e.g., wages and salaries, provisions for impairments, holding gains/losses).
  - Details at the subsidiary level are required to derive accurate estimates of intermediate consumption.

### Flows, stocks, and balance-sheet information for mining firms
- Financial statements limitations:
  - Generally do not contain all information needed to derive transactions and other flows.
  - Cash flow statements and accompanying notes provide some information, but compilers often rely on changes between opening and closing stocks and apply methodologies to derive revaluations and other volume changes (see referred methodology in the Guide).
- Balance-sheet disclosures:
  - Crypto assets appear under categories such as “Digital Valuables,” covering Bitcoin and other crypto assets held and lending denominated in Bitcoins.
  - Liabilities can include borrowings denominated in Bitcoin, with disclosure of outstanding amounts and principal.
  - Limited information is usually available on revaluations of Bitcoin holdings.
- Bitcoin movement statements:
  - Some companies provide Bitcoin movement statements tracking receipts and payments in monetary units and Bitcoins.
  - Receipts examples: Bitcoins received through mining-related rewards and fees, hardware and software sales, proceeds from joint ventures, shareholder loans received in Bitcoin.
  - Payments examples: consulting expenses, salaries, and shareholder loans paid in Bitcoin.
  - Breakup of receipts/payments by residents/nonresidents and by subsidiaries is not available.
- Bitcoin lending/borrowing recording guidance:
  - Compilers should follow the recording guidance in Section G, Chapter 2: for lent/borrowed crypto assets (Bitcoins), only the interest income receivable/payable in Bitcoins should be recorded in the national accounts/balance of payments (as explained in Box 2.8).
  - No transaction in those crypto assets is recorded in the capital account and no changes are made to the balance sheets of the parties involved.

### Compilation guidance and recommended actions for mining companies
- Short-term recommendation:
  - Compilers with access only to group-level financial statements are advised to apply proxy allocation methods to generate country-level estimates from consolidated reports, while documenting assumptions used.
  - Collaboration with international organizations can help harmonize approaches for greater comparability across countries.
- Medium-term recommendation:
  - Collect data from units engaged in crypto mining using standardized templates through targeted surveys.
    - Templates should request country-level data from local subsidiaries on output, input costs, and crypto asset flows/stocks, and a resident/nonresident breakdown of revenues.
    - See the Guide’s Annex 3.2 for a model survey questionnaire.
  - National statistical offices and central banks may need to work closely with regulators, accounting bodies, and companies to encourage improved disclosure practices and reporting to targeted surveys.

---

### Stablecoin issuers — assets, liabilities, flows, and income
- Financial statements coverage:
  - Group-level financial statements typically include information on reserves backing stablecoins and liabilities on account of stablecoins in circulation.
  - Example reserve composition reported: U.S. Treasury bills, other sovereign securities, corporate bonds, repos, money market funds, cash, Bitcoins, and precious metals.
  - Nonfinancial classification:
    - Bitcoins and precious metals are not financial assets in BPM/SNA and should be recorded as part of the nonfinancial balance sheet (should not be recorded in the IIP).
- Classification challenges:
  - Disclosures allow classification of assets into instruments but functional classification (direct/portfolio/other investments) requires assumptions.
  - Residence of counterparties on the asset side is rarely disclosed (e.g., repos and money market fund investments without domicile of issuers).
  - Residence of stablecoin holders is unknown; residency may be approximated using mirror data from major exchanges and custodians or special assumptions (example: assuming overwhelming share of liabilities is held abroad in a case where an issuer is headquartered in a small economy).
- Financial flows from positions:
  - Financial statements show changes in asset and liability levels across reporting periods but do not separate transactions from revaluations nor report by instrument, maturity, or residency.
  - Compilers can derive transactions indirectly from positions by adjusting for revaluation effects using observable market indices (e.g., U.S. Treasury bill indices); Annex 3.3 provides indicative steps using Tether data.
  - Changes in circulating supply of stablecoins can be used to estimate issuance and redemption flows, adjusted for valuation changes.
- Income and expenses:
  - Financial statements often lack detailed income statements; aggregate financial results (net profits/losses) and sometimes dividends are reported.
  - Compilers can reconcile net profit with dividends to deduce undistributed earnings, but allocation of income by instrument, currency, or counterparty is not provided.
  - In absence of detailed data, compilers may apply benchmark yields to reported reserve compositions (e.g., prevailing yields on short-term U.S. Treasury bills) to estimate interest income; aggregate income can be proportionally allocated across instruments by reported shares and observable yields.
- Recommendations:
  - Short-term: combine available disclosures with proxy methods — derive transactions from positions using market indices and exchange rates; apply benchmark yields to estimate income; use mirror or survey data to allocate liabilities by residency.
  - Medium-term: develop standardized reporting templates for stablecoin issuers requiring disclosure of assets and liabilities by instrument, currency, maturity, residency, and income/operating expenses. Such templates could be linked to regulatory initiatives (e.g., GENIUS Act of 2025) and closer cooperation with regulators, supervisors, and exchanges is recommended.

---

### Crypto exchanges and custodians — assets under custody and revenue
- Financial statement coverage and product lines:
  - Group-level financial statements generally cover global operations across product lines and subsidiaries (e.g., custody, buying/selling of crypto assets, staking, lending/borrowing).
- Assets under custody (AOP) disclosures (example: Coinbase):
  - Reports relatively detailed information on assets on platform with breakdown by asset types including Bitcoin, Ether, Solana, USDC, and other crypto assets.
  - Assets under custody may include assets held for institutional clients such as exchange-traded funds; custodial fee revenues are tied to volume of assets under custody.
  - Reports also provide details on crypto assets placed for staking by institutional and retail investors.
- Revenue and expenditure line items disclosed:
  - Revenue items include custody revenue, transaction revenue, blockchain rewards (staking revenue), and stablecoin revenue.
  - Expenditure items include blockchain rewards (staking rewards), blockchain transaction fee, and USDC rewards.
- Data limitations for national statistics:
  - Residency and sectoral allocation of holders are generally not reported (no residency of customers).
    - Example: 83 percent of revenues from custody services are linked to U.S. clients, but this cannot be directly equated with residency of asset holders.
  - No country-level breakdown of non-U.S. activity, preventing bilateral/directional analysis of cross-border flows.
  - No sectoral allocation of clients (institutional vs retail) in many reports.
  - Flow data are not separated into transactions and revaluations, a critical omission given crypto price volatility.
- Usefulness and recommended approaches:
  - Financial statements are a useful starting point for measuring holdings and other items but must be supplemented by other sources and estimations for national accounts and external sector statistics.
  - Short-term guidance:
    - Apply proxies to measure residency and sectoral breakdown of holders (custodial revenue provides a proxy for geographic allocation; trading volume split between institutional and retail may proxy sectoral allocation).
    - Periodically reassess accuracy of proxies/estimates.
    - The Guide provides estimated holdings by residents (USA)/nonresident customers and by sectors for end of 2024 using such proxies (see Annex 3.5).
    - For deriving transactions and revaluations from positions, refer to the numerical example in Box 2.6.
  - Medium-term guidance:
    - Closer collaboration with custodians and regulators and standardized survey questionnaires to collect assets under custody by residence and by institutional sector, and flows separated into transactions, revaluations, and other changes in volume (see Annex 3.6 for a model questionnaire).

*Source: The Guide (sections 3.25–3.43).*

### 3.44 Annex 3.7 provides mapping of key items from the financial statements of mining companies,

### crypto-assets-cg - 3.44 Annex 3.7 provides mapping of key items from the financial statements of mining companies,

### Mapping and use of corporate financial statements for crypto-assets measurement
- Annex 3.7 maps key items from financial statements of mining companies, stablecoin issuers, and crypto exchanges into 2025 SNA/BPM7 categories to assist in measuring crypto assets in national accounts and external sector statistics.
- Corporate financial statements commonly present crypto assets under the item “digital assets”.
- Findings and limitations when using corporate financial statements:
  - Disclosure practices remain inconsistent; many firms report crypto holdings only when material or strategically relevant, and formats vary widely (e.g., MicroStrategy provides detailed quarterly updates; NEXON and Monex Group provide minimal references).
  - Crypto-assets are frequently reported as a single line item without distinguishing types (e.g., Bitcoin versus stablecoins), complicating classification between non-produced assets and financial instruments.
  - Under IFRS, digital assets are commonly valued at cost less impairment, with fair values disclosed in notes; US GAAP requires fair value reporting at the end of the reporting period.
  - Financial statements often lack counterparties and geographic detail, making cross-border positions in balance of payments or international investment position statistics difficult to assess (e.g., a U.S. firm may hold stablecoins issued by a Cayman-based entity, but this is rarely evident).
- Example reported holdings as of December 31, 2024:
  - MicroStrategy reported digital asset holdings (primarily Bitcoins) valued at $23.9 billion.
  - Tesla reported digital asset holdings (primarily Bitcoins) valued at $1.076 billion.
- Corporate reports are useful but should be complemented by other data sources for comprehensive coverage.

### Institutional holdings and ETFs
- Crypto exchange-traded funds (ETFs) have become significant institutional holders of Bitcoin and other crypto-assets.
  - ETFs disclose holdings in regulatory filings and issuer reports, often noting custody in regulated cold storage with institutional custodians (e.g., Coinbase Custody, Fidelity Digital Assets).
  - Following the launch of spot Bitcoin ETFs in the United States in early 2024 and similar ETPs in Europe, U.S. funds such as BlackRock iShares Bitcoin Trust, Fidelity Wise Origin Bitcoin Fund, and Grayscale Bitcoin Trust collectively hold large volumes of Bitcoin—together amounting to over one million BTC equivalent by end-2025, according to publicly disclosed ETF reports and SEC filings.
  - European crypto ETPs issued by providers such as CoinShares and 21Shares also hold substantial Bitcoin and Ether positions, though on a smaller scale.

### Data collection from enterprises and households
- Enterprise surveys:
  - Targeted questions on crypto assets in enterprise surveys are important to capture flows/stocks, revenue, and expenditures of firms engaged with crypto.
- Household sector:
  - Households lack comparable administrative or accounting records; targeted household surveys are often the only direct source for household holdings and transactions in fungible crypto assets and NFTs.
  - Annex 3.8 provides model questions for household and enterprise surveys covering flows and stocks.

### Commercial (on-chain analytics) data sources and limitations
- Commercial blockchain analytics firms (e.g., Chainalysis, Elliptic, Glassnode) provide datasets and attribution techniques to identify wallet clusters, categorize user types, and estimate flows between entities and geographies.
- Core data products from on-chain analytics companies:
  - Market capitalization and market prices of different crypto assets.
  - On-chain transaction data, including volume, direction (inflows/outflows), and timing.
  - Geographic estimates of crypto activity including main counterpart economies of inflows/outflows.
  - Entity-level, often heuristics and model-based classification (e.g., exchanges, illicit actors, DeFi protocols).
  - Wallet balances, historical and current, across major blockchains.
  - Market intelligence dashboards with cross-border movement estimates and risk scores.
- Key limitations (Box 3.3) when using commercial on-chain data for BPM7/2025 SNA macroeconomic statistics:
  - On-chain data capture only part of crypto activity; substantial activity occurs off-chain (e.g., through centralized crypto exchanges or custodial providers).
  - Blockchain data record movements between addresses, not ultimate beneficial owners; on-chain flows may include internal transfers (e.g., hot to cold wallets) rather than changes in economic ownership across borders.
  - Databases rely on assumptions and approximations that may not meet BPM7/2025 SNA methodological requirements; residency inference is challenging. Three common residency inference approaches:
    - Geographical location of the crypto exchange registration (after matching blockchain addresses with exchanges).
    - Location of web visitors to the crypto exchange website and geographic distribution of app usage.
    - Entry and exit fiat currency as a proxy of residency of units engaged in crypto transactions.
  - Developing on-chain analytics requires specialized IT and machine-learning capabilities; national statistical offices and central banks may build capacity in-house or rely on commercial providers, both entailing additional costs and partial disclosure of commercial methodologies.
- Usefulness of commercial data for macro trends:
  - Commercial data can help attribute macroeconomic trends to crypto flows; example cited in an IMF Article IV Report for Vietnam where Chainalysis data suggested a relationship between rising unrecorded crypto transactions and a large increase in the balance of payments statistical discrepancy:
    - Statistical discrepancy increased from less than US$10 billion before 2022 to averaging US$28 billion during 2022–24.
  - At the time of drafting the chapter (June 2026), Chainalysis covers about 90% of daily market volume across 120 digital assets including seven crypto assets: Bitcoin (BTC), Bitcoin Cash (BCH), Ether (ETH), Litecoin (LTC), Tether (USDT), TrueUSD (TUSD), USD Coin (USDC). Chainalysis monitors transactions of 800 exchanges.

### Data sources for NFTs
- Potential sources and strengths/limitations:
  - Tax data and financial reports:
    - Tax filings capture realized gains and losses from NFT sales (e.g., in the United States), enabling estimates of total value of NFT transactions by residents over a given period.
    - Reporting of holdings for capital gains purposes provides partial information on resident ownership of NFTs when combined with survey data.
    - Financial reports of large NFT creators are often not publicly available (e.g., Yuga Labs), limiting use.
    - Total value of NFTs sold can be observed via marketplaces or blockchain data, but buyers are largely anonymous and intermediate costs or operational expenditures are unavailable.
    - Crypto exchanges (e.g., Coinbase) report “assets on platform” for fungible crypto assets; NFTs are generally excluded or aggregated without differentiation.
  - Crypto exchanges and custodians:
    - Exchanges and custodians maintain internal records of customer NFT holdings, including residency and KYC details; if accessible, such data could allow estimation of NFT stocks by resident sectors.
    - See Section 6, annex 3.6 for sample questions on NFTs for collecting related flows and stocks from crypto exchanges.
  - Surveys of creators and potential holders:
    - Targeted surveys of NFT creators can capture output, intermediate consumption, and export share of created NFTs.
    - Adding NFT questions to household and enterprise surveys helps capture flows/stocks; Annex 3.9 provides a model questionnaire for creators; Annex 3.8 (Section 2) provides sample questions for household and enterprise surveys.
  - Blockchain and marketplace data:
    - All NFT transactions and holdings are recorded publicly on blockchains, offering near-complete visibility of market activity (e.g., CryptoPunks have fully traceable histories).
    - Blockchain and marketplace data (accessible through analytics firms like Chainalysis or platforms such as OpenSea) provide transaction volumes and timestamps.
    - Tracing wallet addresses to specific countries or institutional units remains nearly impossible; blockchain records alone are insufficient for compiling NFT-related flows and stocks consistent with national accounts and external sector statistics.
- Outlook:
  - Comprehensive data categorized by the three types of NFTs (as envisaged in BPM7 and the 2025 SNA) may be achievable in the medium term with continued coordination among creators, statistical agencies, regulators, tax authorities, and custodians.

### DGI-3 Recommendation 11: data collection templates and related guidance
- Recommendation 11 establishes a data collection framework covering CBDCs, stablecoins, and other crypto assets used as a means of payment.
- The Rec. 11 task team (Rec. 11 TT) agreed on data collection templates (available on the G20 DGI Recommendations website) and conducted test data collection during July 15 to December 15, 2025.
- Templates and their aims:
  - Template for Central Bank Digital Currency (CBDC): Data on issuance of CBDCs by central banks and monetary authorities and the holdings of the CBDCs.
  - Template for Stablecoins Issuers: Data on issuance of stablecoins by resident entities and the holdings of stablecoins.
  - Template for Holdings and Flows of Crypto Assets: Data on crypto assets and stablecoins held by the customers of resident crypto exchanges and custodians.
- Template-specific guidance and compilation considerations:
  - Stablecoin Issuers template:
    - Captures stablecoin issuance by resident entities and their holders, requiring holder data by institutional sectors: households, central bank, deposit-taking corporations, other financial corporations, money market funds, general government, nonfinancial corporations (split into private and public NFCs).
    - Nonresident holders: data required by country and broad institutional sectors in total and each country.
    - Residence of stablecoin holders is often unknown from financial reports; residency may be approximated using mirror data from major exchanges and custodians or by other assumptions (e.g., assuming overwhelming share held abroad as with Tether headquartered in El Salvador).
    - Example: Coinbase financial report provides holdings of different stablecoins by US residents on its platform.
    - In the medium term, a model survey questionnaire for stablecoin issuers (Annex 3.4) will aid reporting.
  - Holdings and Flows of Crypto Assets template:
    - Captures data from crypto exchanges and custodians on crypto assets held by customers; structured in two parts: (a) Data by broad categories of crypto assets; (b) Data on crypto-by-crypto.
    - Data by broad categories requires reporting flows and positions (opening position, inflows/outflows, other changes, and closing position) of crypto assets (total, and breakdown into stablecoins and unbacked crypto assets) under custody of resident custodians belonging to resident and nonresident holders; resident holders reported by institutional sectors as per the stablecoin template.
    - Crypto-by-crypto requires reporting holdings in seven crypto assets under custody of resident custodians belonging to resident and nonresident holders: Bitcoins, Ether, Tether, BNB, SOL, USDC, and XRP.
    - Short-term compilation: use guidance in section C.3 and Annex 3.5 to compile custody-based holdings from financial reports of custodians; example: Coinbase reports assets under custody with breakdown by major crypto asset types (e.g., Bitcoin, Ether, Solana, USDC).
    - Flows are not generally available from financial reports and should be estimated following Annex 3.3; estimates should account for crypto asset loss, theft, and seizures where relevant.
    - Additional sources: financial reports of custodians, mining companies, reporting corporations (e.g., MicroStrategy and Tesla for the USA), regulatory data, and tax data (mainly flows) can inform resident institutional sector holdings.
    - For cross-border inflows and outflows, ITRS (where operational, e.g., Brazil) could be an important source.

*Source: IMF — crypto-assets-cg (sections 3.44–3.70 and annex references).*

### 3.71 To achieve comprehensive reporting of these templates (especially with regards to positions) in

### crypto-assets-cg - 3.71 To achieve comprehensive reporting of these templates (especially with regards to positions) in

### Data collection and reporting templates
- The chapter recommends collecting data from custodians—domestically headquartered exchanges and local branches or subsidiaries of foreign-owned platforms—using the standardized questionnaire (Annex 3.6) to achieve comprehensive reporting of templates, especially for positions.
- The templates translate into reporting data on crypto assets by SNA institutional sectors for:
  - crypto assets without a corresponding liability designed to function as a medium of exchange (AN22), and
  - crypto assets with a corresponding liability designed to serve as a general medium of exchange but not issued or authorized by monetary authorities or government (AF22).
- Distinction emphasized:
  - The templates focus on crypto holdings/flows of resident institutional sectors with resident custodians.
  - Institutional sector accounts encompass holdings/flows of resident sectors with both resident and nonresident custodians.

### Cross-border nature and need for international data sharing
- The chapter finds that relying solely on national data sources (administrative records or surveys of CASPs and households) can be insufficient because residents often transact with or hold assets through service providers abroad.
- Examples cited:
  - A household survey in Country A may understate holdings if residents do not report holdings with nonresident exchanges and custodians.
  - Tax filings may not capture crypto assets kept in custodial wallets outside the country.
- Concentration of activity:
  - Crypto asset activity is concentrated among a small number of global players (large exchanges, major stablecoin issuers, and custodians) that operate from a few economies but provide services globally.
  - Without information from foreign providers, national compilers may underestimate residents’ holdings, cross-border flows, and related income.
- Recommendation:
  - International organizations should facilitate and coordinate cooperation by creating frameworks for standardized reporting and enabling exchange of data on positions, holders, and residency, while acknowledging legal constraints and promoting workable arrangements to strengthen global statistical coverage and macroeconomic statistics quality.

### Measurement challenges specific to crypto assets
- New features complicate measurement: decentralized issuance, pseudonymity, and digital custody.
- Conventional data sources and collection mechanisms cannot fully capture the scope of activities because of these features.

### Models for coordinated data collection
- The chapter suggests internationally coordinated databases or data-sharing agreements modeled on the Coordinated Portfolio Investment Survey (CPIS) or Coordinated Direct Investment Survey (CDIS) to provide countries with reliable information on crypto asset holdings and flows relevant to their residents, even when service providers are abroad.
- Effective data collection arrangements can be developed with major ecosystem actors—stablecoin issuers, large exchanges, and mining companies.

### Model survey questionnaires (annex summaries and guidance)
- Annex 3.2: Model survey questionnaire for crypto mining companies
  - Guidance: report all values in USD or domestic currency; use the price on the day of the transaction when reporting value of Bitcoins mined; indicate “Not Applicable (N/A)” where appropriate; provide best estimates if exact figures unavailable; ensure consistency across reported figures.
  - Key items to collect include: number of Bitcoins mined (Units), value of Bitcoins mined (USD), explicit fees in Bitcoins and USD, input costs (USD) with breakdowns (electricity, maintenance, remuneration, other), selected balance sheet items (Currency and deposits USD; Bitcoin USD; Gross Capital Formation USD; Loans USD; Debt securities USD; Equity USD), revaluations and other changes in volume (USD), and detailed uses of Bitcoins (sold to resident and nonresident exchanges—Units and USD).
- Annex 3.4: Model survey questionnaire for stablecoin issuers
  - Guidance: report monetary values in USD or domestic currency; use end-of-period market values for positions and transaction values for flows; clearly distinguish resident and nonresident counterparties; indicate methods for estimates.
  - Key items to collect include: total value of stablecoins issued (by currency), total value of stablecoins outstanding, breakdown of stablecoin holders by residency (example format: provide estimated breakup, e.g., 80 and 20 percent), breakdown of resident holders by domestic sectors, assets of the stablecoin issuer with breakdowns (Currency and deposits; Debt securities; Loans; Equity; Money Market Funds; Reverse repurchase agreements; Repurchase agreements; Other financial assets), nonfinancial assets (Crypto assets without a corresponding liability; Gold and other precious metals), and revenues/expenditures with “of which from nonresidents” for each relevant line.
  - Note: stablecoin issuer questionnaires should report “of which” claims on nonresidents for each financial asset and identify relevant functional categories.

### Deriving transactions and price changes from positions (Tether example)
- Annex 3.3 provides indicative steps to derive transactions and price changes from positions using Tether as an example:
  - Table 1 (Tether: Composition of Assets) reports Amount (USD) by asset category:
    - Cash & Cash Equivalent & Other Short-Term Deposits: 118,335,279,417 (As of Dec 31, 2024) and 121,647,076,498 (As of March 31, 2025)
      - U.S. Treasury Bills: 94,471,651,607 and 98,523,657,338
      - Overnight Reverse Repurchase Agreements: 14,101,623,321 and 15,093,981,718
      - Term Reverse Repurchase Agreements: 3,077,452,467 and 1,613,610,695
      - Money Market Funds: 6,506,444,067 and 6,285,638,008
      - Cash & Bank Deposits: 108,844,601 and 64,302,555
      - Non-U.S. Treasury Bills: 69,263,354 and 65,886,184
    - Corporate Bonds: 14,270,773 and 14,354,777
    - Precious Metals: 5,318,875,241 and 6,663,205,657
    - Bitcoin: 7,857,529,277 and 7,662,210,442
    - Other Investments: 3,984,793,433 and 4,462,144,209
    - Secured Loans: 8,194,007,406 and 8,825,524,405
    - Total (1+2+3+4+5+6): 143,704,755,547 and 149,274,515,988
  - Price movement and transaction estimation:
    - The price of US treasury bills increased by 3 percent during the quarter; the index would be 103 for end-March against 100 at the end of December 2025. Applying the adjusted value for end-March data, the transactions are estimated at USD 1,200 million.
    - For Bitcoin holdings: price decreased from $93,812 at end-December 2024 to $82,704 at end-March 2025; the average price would be $88,258.
    - Holdings in bitcoins: increased from 83,758 BTC (calculated as USD 7,856 million ÷ $93,812 per BTC) to 92,646 BTC (calculated as USD 7,662 million ÷ $82,704 per BTC), implying an increase of 92,646−83,758=8,888 BTC.
    - Multiplying the BTC increase by average price: 8,888×88,258≈784 million USD (amount representing purchase of Bitcoins; positions reflected in the nonfinancial balance sheet in national accounts).

### Illustrative statistics and seizure examples
- Selected seizures of crypto assets (arranged by value of assets seized) reported in Annex 3.1 include:
  - USA — Bitcoin — $3.6 billion — Bitfinex Hack (2022). Department of Justice seized funds laundered from 2016 Bitfinex hack.
  - USA — Bitcoin — $3.36 billion — Silk Road (2021). Seizure of Bitcoin stolen from Silk Road darknet marketplace.
  - United Kingdom — Bitcoin — $3 billion — Money laundering operation.
  - Germany — Bitcoin — $2.55 billion — Piracy website and laundering after Bybit hack.
  - USA — Crypto assets — $225.3 million — Crypto Investment Scams (2025)—Largest seizure in U.S. Secret Service history; linked to fraud and laundering.
  - Spain — Crypto assets — EUR 27 million — Transnational criminal organization.
  - Albania — Crypto assets — $10 million — Organized crime group.
  - Netherlands — Crypto assets — $7.6 million — Illicit exchanges shut down.
  - Australia — Bitcoin — $2.6 million — Stolen from French crypto exchange.
  - Israel — Crypto assets — $1.7 million — Linked to Hezbollah and Iran's Quds Force.
  - Serbia — Crypto wallets — $1 million — Fake crypto investment schemes.
- Estimated crypto holdings by USA and non-USA residents (with Coinbase) as of December 31, 2024 (Annex 3.5):
  - Total Assets Under Custody: 404.0 ($bn)
    - Bitcoin: 235.4 ($bn) — US Residents (83%) 195.4 ($bn); Non-US (17%) 40.0 ($bn)
    - Ether: 54.2 ($bn) — US Residents (83%) 45.0 ($bn); Non-US (17%) 9.2 ($bn)
    - Solana: 21.3 ($bn) — US Residents (83%) 17.7 ($bn); Non-US (17%) 3.6 ($bn)
    - Other: 87.1 ($bn) — US Residents (83%) 72.3 ($bn); Non-US (17%) 14.8 ($bn)
    - USDC: 6.1 ($bn) — US Residents (83%) 5.1 ($bn); Non-US (17%) 1.0 ($bn)
  - Total split: 404.0 ($bn) — US Residents 335.3 ($bn); Non-US 68.7 ($bn)
- Estimated crypto holdings by resident institutional sectors (with Coinbase) as of December 31, 2024:
  - Total AUC for resident clients: 335.3 ($bn)
  - Sectoral allocation proxy used: Financial Corporations (80%) and Households (20%)
    - Bitcoin: 195.4 ($bn) — Financial Corporations (80%) 156.3 ($bn); Households (20%) 39.1 ($bn)
    - Ether: 45.0 ($bn) — Financial Corporations 36 ($bn); Households 9 ($bn)
    - Solana: 17.7 ($bn) — Financial Corporations 14.1 ($bn); Households 3.5 ($bn)
    - Other Crypto: 72.3 ($bn) — Financial Corporations 57.8 ($bn); Households 14.5 ($bn)
    - USDC: 5.1 ($bn) — Financial Corporations 4.1 ($bn); Households 1 ($bn)
  - Total: 335.3 ($bn) — Financial Corporations 268.4 ($bn); Households 67.1 ($bn)
  - Note: These are illustrative calculations and subject to limitations; custodial revenue and trading volume splits used as proxies (83 percent geographic allocation to U.S. residents; institutional trading volume about 81 percent leading to an 80/20 asset split between financial corporations and households).

*Source: IMF publication on compiling crypto-asset statistics (crypto-assets-cg).*

### Annex 3.6. Model Survey Questionnaire for Crypto Exchanges

### Annex 3.6. Model Survey Questionnaire for Crypto Exchanges

### Purpose and scope
- The questionnaire helps in collecting data from resident crypto exchanges/custodians to support the compilation of items related to them (including the crypto holdings of their customers) following the BPM7 and 2025 SNA frameworks.  
- The survey scope encompasses domestically headquartered exchanges and local branches or subsidiaries of foreign-owned exchanges.

### Questionnaire structure (major sections)
- Section 1: General Information
  - Legal name of the reporting entity
  - Country of incorporation and registration
  - Contact person (name, title, email, phone)
  - Reporting period (start and end dates)
  - Nonresident ownership questions: “Does this enterprise have a nonresident owner? Yes/No.” and “If the answer is Yes, does the nonresident owner hold 10 percent or more equity? Yes/No.” (If Yes and relevant, provide further details in Section 2 as the Stablecoin issuer is a direct investment enterprise)
- Section 2: Shareholding Structure of the Enterprise (Equity liabilities of the crypto exchange)
  - Reporting template includes: Name and country of residence of shareholder; Opening position (end of year t-1 in $)/Percentage of total; Financial transactions during year t (Acquisitions, Disposals); Other changes during year t; Closing position (end of year t in $)/Percentage of total.
- Section 3: Assets/Liabilities and related transactions and other changes (USD)
  - 3A. Assets of the Exchanges (USD): Position reporting (Opening, Net Financial Transactions, Revaluations, Other changes in volume, Closing) for:
    - 3.1 Currency and deposits (Of which claims on nonresidents)
      - 3.1.1 Stablecoins
      - 3.1.2 Customer custodial funds
    - 3.2 Accounts receivable
    - 3.3 Other financial assets (please specify)
    - Nonfinancial assets:
      - 3.4 Crypto assets without a corresponding liability (e.g., Bitcoins)
      - 3.5 other nonfinancial assets (specify)
  - 3B. Liabilities of the Exchanges (USD): Position reporting for:
    - 3.6 Currency and deposits (Of which payable to nonresidents)
      - 3.6.1 Customer custodial fund liabilities
    - 3.7 Other accounts payable
    - 3.8 Other liabilities (please specify)
- Section 4: Revenues and Expenditures (Year t) (USD)
  - Revenues: 4.1 Transaction revenue (Of which from nonresidents), 4.2 Blockchain rewards (staking revenue), 4.3 Custodial fee, 4.4 Stablecoin revenue, 4.5 Other subscription and services revenue, 4.6 Interest (4.6.1 Interest from crypto lending), 4.7 Dividends, 4.8 Other items (specify)
  - Expenditures: 4.9 Blockchain rewards (staking rewards), 4.10 Blockchain transaction fee, 4.11 USDC reward, 4.12 Other expenses (e.g., rental, legal, IT, marketing, payment processing), 4.13 Remuneration of employees, 4.14 Interest (4.14.1 Interest on crypto borrowing), 4.8 Distributed dividends, 4.10 Other items (specify)
- Section 5: Assets under custody — fungible crypto assets (USD)
  - Itemized by crypto asset: 5.1 Bitcoins, 5.2 Ether, 5.3 Tether, 5.4 BNB, 5.5 SOL, 5.6 USDC, 5.7 XRP, 5.8 Other stablecoins
  - Totals and “Of which belong to nonresidents” breakdowns:
    - Total 1 (Crypto assets without a corresponding liability): 5.1+5.2
    - Total 2 (Crypto assets without a corresponding liability): 5.3+...+5.8
    - Total (Total 1 + Total 2) and corresponding nonresident breakdowns
- Section 6: Assets under custody — NFTs (USD)
  - Itemized top collections: 6.1 Crypto Punks, 6.2 Bored Ape Yacht Club, 6.3 Mutant Ape Yacht Club, 6.4 Pudgy Penguins, 6.5 Autoglyphs
  - “Of which belong to nonresidents” and total of the listed collections
- Additional Information / General Guidelines
  - Report all monetary values in USD or domestic currency.
  - Use end-of-period market values for positions and transaction values for flows.
  - If exact figures are unavailable, provide best estimates and indicate that they are estimated, and indicate the method used.
  - Clearly distinguish between resident and nonresident counterparties.
  - Ensure consistency between assets and liability positions and flows.
  - At the initial stage, collecting data with a resident/nonresident breakdown may be sufficient; compilers may include main counterparties if capacity exists.
  - The survey scope for enterprise and household surveys encompasses total stock and flows of crypto assets, including assets held in self-hosted and cold wallets.

### Key definitions and notes (selected)
- Customer custodial funds: restricted cash and cash equivalents maintained in segregated accounts of the Company at financial institutions and asset managers that are held for the exclusive benefit of customers and deposits in transit from payment processors and financial institutions.
- Crypto assets without a corresponding liability: e.g., Bitcoins and other crypto assets generally part of the assets of exchanges.
- Customer custodial fund liabilities: obligation to return cash deposits held by customers in their fiat wallets and unsettled fiat deposits and withdrawals.
- Transaction revenue: fee/commission earned for facilitating a buy, sell, or convert transaction for a customer; includes explicit fees and profit from spread; generally available by crypto asset.
- Stablecoin revenue: example provided specific to Coinbase; refers to Coinbase’s share of income derived from its partnership with Circle for the promotion and distribution of the USDC stablecoin. A clarification note on the recording of stablecoin revenue and USDC rewards in macroeconomic statistics was agreed with the AEG and BOPCOM at the October 2025 meetings.
- Staking rewards: amount of staking rewards distributed to customers who have staked their crypto assets through exchanges.
- Blockchain transaction fees: network fees paid to miners or validators when an exchange (on behalf of its users) sends transactions on a blockchain.
- USDC reward: expense specific to Coinbase representing the portion of stablecoin revenue paid out to users.

### Enterprise and household survey modules (Annex 3.8) — selected items
- Fungible crypto assets:
  - Opening stock by type (units and value in USD): Stablecoins (AF22); crypto assets without a corresponding liability designed to function as a medium of exchange (AN22)
  - Crypto assets purchased during year t (type, units and value in USD); Of which purchased from nonresident exchanges (USD)
  - Closing stocks by type (end of Year t in USD)
  - Crypto assets lost due to theft/hacking/seizures (type, units and value in USD)
  - Revenue from lending and staking of crypto assets (USD) and “Of which from nonresidents”
  - Income from mining (households): number and value of crypto assets mined plus transaction fee received (USD); Of which sold to nonresident exchanges (USD)
  - Expenditures related to mining and borrowing (USD) and “Of which paid to nonresidents”
- NFTs:
  - Opening stocks by type (beginning of Year t in USD)
  - NFTs purchased during year t (type, units and value in USD); Of which purchased from nonresident marketplaces/creators (USD)
  - Closing stocks by type (end of Year t in USD)

### Mapping to BPM7 / 2025 SNA and transaction recording (selected illustrations)
- The mapping table links financial statement items involving crypto assets to BPM7 and 2025 SNA classifications across current account, capital account, financial account, and production/earnings accounts. Examples:
  - Bitcoin received as mining reward: Current account → Services → Computer services (credit) + Capital account → Acquisition of nonproduced nonfinancial assets (debit); Production account (output of mining); Capital account (acquisition less disposals of nonproduced nonfinancial assets)
  - Bitcoins realized for cash (sold for fiat): Capital account → Disposal of nonproduced nonfinancial assets (credit) + Financial account → Other investment → Currency & deposits (NAFA ↑)
  - Bitcoin payables/receivables for cross‑border trade settled in Bitcoins: Other investment → Other accounts payable/receivable – Trade credits; settlement via Capital account (disposal/acquisition of Bitcoins); Financial account → Other accounts payable/receivable; Capital account for BTC settlement
  - Stablecoins with a claim on the issuer: treated as financial assets (Currency and deposits) of the owner/original holder if they are external assets of the owner.

### Country experience — Brazil (Central Bank of Brazil, BCB) — data collection approach and 2025 results
- Data collection framework:
  - The Central Bank of Brazil (BCB) compiles crypto assets statistics with data sourced from foreign exchange transaction contracts settlements (i.e., an ITRS). Financial institutions conducting foreign exchange settlements register purchase and sale transactions of foreign currency; recorded information includes transaction nature (from a list with approximately 100 foreign exchange codes), transaction value, and identification of resident and nonresident parties. Two foreign exchange codes exist for recording transactions with crypto assets. An “Other Specifications” free-text field allows identification of the crypto asset name (e.g., USDT or BTC).
  - Financial institutions transmit foreign exchange contracts to BCB in XML files monthly; records are processed, stored, loaded into a balance of payments data warehouse (DW), and allocated to specific balance of payments accounts based on predefined rules, transaction type, foreign exchange code, and resident institutional sector. Data are validated and may be altered to better reflect the economic nature of transactions during compilation.
- 2025 aggregate figures (based on ITRS data):
  - Total purchases: US$ 18.3 billion
  - Total sales: US$ 0.8 billion
  - Purchases of crypto assets with corresponding liability: US$ 17.3 billion
  - Sales of crypto assets with corresponding liability: US$ 0.4 billion
  - Purchases of crypto assets with corresponding liability represented 97 percent of the net purchases in 2025.

*Source: IMF — Annex 3.6. Model Survey Questionnaire for Crypto Exchanges; related annexes and Chapter 4 country experiences.*

### 4.6 Foreign exchange contacts related to crypto assets are extracted by a text reading and

### crypto-assets-cg - 4.6 Foreign exchange contacts related to crypto assets are extracted by a text reading and processing algorithm, which identifies the specified crypto assets and the value associated with each crypto asset

### Brazil (BCB): extraction algorithm, reference list, and reporting reforms
- Text-reading and processing algorithm:
  - Identifies specified crypto assets and the value associated with each crypto asset.
  - Compares the crypto asset to a reference list and categorizes the transaction value as with or without corresponding liability.
  - If the crypto asset is not found in the list, verification is made to determine which category it fits and it is then added to the list.
  - Verification of the category of each crypto asset is conducted through internet searches to determine whether they have corresponding liability or not.
- Reference list (Table 4.1) — Code / Name / With or without corresponding liability:
  - USDC / USD Coin / With corresponding liability
  - BTC / Bitcoin / Without corresponding liability
  - USDT / Tether / With corresponding liability
  - ETH / Ether / Without corresponding liability
- Data coverage and monitoring:
  - In the last quarter of 2025, the names of the currencies had been recorded in approximately 80 percent of the crypto asset purchase records and 67 percent of the sale records.
  - BCB’s goal is, through monitoring activities of the records, to reach 100% for both purchase and sale.
- Complementary sources and limitations:
  - The Federal Revenue of Brazil publishes data on transactions with crypto assets declared by residents, but without determining whether the transaction is domestic or between a resident and a nonresident.
  - The Federal Revenue data allows verifying which types of currencies are being transacted and the evolution of crypto asset use in Brazil but does not constitute a direct source for residency breakdowns.
- New reporting procedures and timeline:
  - New data reporting procedures for resident Virtual Asset Service Providers (VASPs) are expected to be received by BCB as of June 2026.
  - In November 2025 BCB started to regulate crypto assets market in Brazil and this new entity, VASP, is part of the framework.
  - Reporting procedures for resident VASPs require cross-border transactions – from customers and on their own - involving payments, transfers, or exchanges of crypto assets to be reported to the BCB.
  - Required reporting details are identical to the ITRS, including information on related parties, type of crypto asset, and nature of the transaction.
  - Data should be reported on a transaction-by-transaction basis, without a threshold, and included in the balance of payments on a monthly basis.
  - After the first data collection, plans include additional procedures for VASPs to report crypto asset positions by type of asset.

### Estonia: sector evolution, integrated reporting, data items, and key 2024 observations
- Sector evolution and regulatory context:
  - Since 2018, Estonia’s crypto sector expanded rapidly; peak in 2019 with more than 1,200 VASPs operating in the country.
  - Parliamentary amendments at the end of 2019 tightened standards (physical presence, payment accounts, minimum share capital, customer identity verification).
  - Further amendments in force in 2022 imposed higher capital requirements, mandatory external audits, and stricter board conditions.
  - In 2023 the Ministry of Finance introduced regulations requiring VASPs to submit regular reports on their activities, effective from January 1, 2024.
  - By the end of 2023, the number of licensed service providers declined to around 50.
- Integrated reporting framework and quarterly reports:
  - VASP reports stipulated by the Ministry of Finance are integrated reports used for supervisory purposes by the FIU and for compilation of official statistics by Eesti Pank.
  - Report on General Activity: location, websites, brands, acceptance of cash payments, number of crypto ATMs, number of customers, number of customers subject to due diligence measures; providers’ websites help identify service packages.
  - Balance sheet items disaggregated by residency—residents and nonresidents; income statements generally without residency breakdown except for sales revenue and goods/raw materials/services (nonresident totals separately reported).
  - Report on Customers’ Assets: quarterly information on customers’ asset holdings, disaggregated by type of asset, type of customer, and residency; assets classified as fiat currency; major crypto assets among the global top ten by market capitalization (Bitcoin, Ether, BNB, Cardano, and Tether); privacy-enhancing crypto assets (Monero, Zcash, Dash); crypto assets issued by service providers themselves; other crypto assets. Customers classified as banks, other financial institutions, other legal entities, natural persons, and politically exposed persons.
  - Report on Virtual Asset Services Provided: quarterly turnover data by type of service, type of customer and residency, type of ultimate beneficiary, number of customers, number of transactions, and value of transactions. Services classified as:
    - exchange between fiat currency and crypto assets (and vice versa);
    - exchange between different crypto assets;
    - services related to organizing offers or sales and related financial services;
    - transfer services; and wallet services.
  - Ultimate beneficiaries classified as residents, nonresidents, or mixed. Turnover data cannot be disaggregated by type of asset.
- Key observations based on 2024 crypto assets data (Box 4.1):
  - Number of service providers' customers varied between 1.4 and 1.6 million.
  - Out of 45-50 service providers, only 3-5 companies had more than 100,000 customers.
  - Crypto assets of customers exceed 2 billion euros, of which around 98 percent belonged to nonresident customers.
  - Over 90 percent market share was formed by four companies, with nonresident customers from specific regions.
  - Customer composition: Over 85 percent of customers are natural persons while 10–15 percent are other legal persons.
  - Asset shares by type:
    - Bitcoin represents one-third of customers total assets.
    - Ether represents 15 percent of customers total assets.
    - Stablecoin Tether forms around four percent and is used mainly by companies from off-shore countries.
    - Privacy-enhanced crypto assets and assets issued by the service providers have very small share.
    - Other crypto assets have shares between 35-40 percent.
  - Service and flow observations:
    - Services providing crypto assets against fiat currency have higher proportion compared to other service types.
    - High turnovers of transfers and wallet services compared with net buying and selling figures and customer asset positions may indicate flows of crypto to customers’ non-custodial wallets outside the scope of surveys.
    - Majority of actual beneficiaries for turnover by Estonian resident companies are nonresidents.
    - Reporting caveats: types of services can be double-recorded and may not show direction; wallet services and transfers can be mixed; revenue from staking can be recorded under wallet services and other services; direction for staking activities (main amounts, crypto borrowing) is not detectable.

### Germany: direct reporting system, classification codes, decision tree, data quality, and planned improvements
- Direct reporting mechanism and threshold:
  - Direct reporting mechanism mandates domestic sectors to report international transactions; threshold for reporting starts at payments of more than €50,000.
  - The German reporting system is governed by the Foreign Trade and Payments Act (Außenwirtschaftsgesetz - AWG) and the Foreign Trade and Payments Regulation (Außenwirtschaftsverordnung - AWV), adheres to EU regulations and international trade agreements, and since 2025 explicitly includes transactions involving crypto assets.
- Required transaction information:
  - Reporting agents must provide detailed information for each transaction containing crypto assets including the value of the transaction, the direction of payment (inflow or outflow of cryptos), and the purpose of the transaction.
  - Other relevant information that may assist classification and analysis should also be included.
- Treatment of mining and swaps:
  - Mining of crypto assets, such as Bitcoin, is reported under computer services.
  - Swaps of crypto assets are taken into account; reporting agents are asked to report both sides of swaps.
- Four new reporting codes (paragraph 4.21) and their broad classifications:
  - I. Crypto assets that do not have a corresponding liability — categorized as nonfinancial assets.
  - II. Foreign crypto assets that are associated with corresponding liabilities — categorized as financial assets.
  - III. Domestic crypto assets that also have corresponding liabilities — categorized as financial assets.
  - IV. Nonfungible tokens (NFTs) — classified as nonfinancial assets.
- Positions and compilation approach:
  - Positions in crypto assets relevant for the international investment position (IIP) do not have to be reported directly.
  - For nonfinancial crypto assets this has no consequence since they are not part of the IIP.
  - For financial crypto assets, positions are calculated by cumulating transactions.
  - Bundesbank regards cumulation of transactions as acceptable for now given the current low volume of transactions in financial crypto assets relative to the overall IIP and less volatility of some financial crypto assets (e.g., certain stablecoins).
- Data quality management:
  - Plausibility checks of reports assess the plausibility of reported amounts or descriptions in combination with the direction of the transaction or other issues.
  - Communication with reporting agents can resolve cases, though for efficiency reasons this is done only for transactions above a certain threshold.
- Decision tree and assumptions for classification (paragraphs 4.24–4.26):
  - Step 1: Identify the geographical location of the counterpart. Preferably use counterpart country of residence; if unknown, use the country of residence of the trading venue as second-best solution.
    - If the counterpart is assessed as domestic, no external transaction reporting is required.
    - If neither counterpart location nor trading venue location can be identified, assume the counterpart is located abroad (Assumption I) and use country code W19 (not allocated geographically).
  - Step 2: Determine type of crypto asset via checks for four types specified in the new codes.
    - First check if the asset is an NFT.
    - If not an NFT, check if the asset has a corresponding liability (issuer may serve as indicator).
      - If no corresponding liability, use nonfinancial crypto assets code.
      - If corresponding liability exists, classify as financial crypto asset and determine domestic or foreign issuer.
      - If issuer location cannot be identified, assume the crypto asset is foreign (Assumption II) and use foreign financial crypto asset code.
- Reporting and compilation challenges:
  - Crypto market dynamism leads to rapid appearance/disappearance of business models, crypto assets, sudden changes in counterparties’ information on trading platforms, and designed (pseudo-)anonymity that complicates counterparty/country identification.
  - Mining activity contributes to higher amounts in W19.
  - Implementation requires training of compilers and comprehensive information for reporting agents.
  - Specific limitations of the German system:
    - Recording the use of crypto assets for payment of goods or services is not always possible.
    - Current system does not facilitate precise identification of the type of crypto asset being traded beyond the four general types.
    - No direct collection of stocks of crypto asset (relevant for financial crypto assets).
    - System does not allow separate identification of digital versions of conventional financial assets.
    - Different types of NFTs cannot be distinguished; all NFTs are treated as NFTs with limited ownership rights to a specified asset or commodity that go beyond personal use.
- Planned adjustments and future directions:
  - Medium term: implement a coding system to identify specific type of crypto asset within a transaction (e.g., Digital Token Identifier).
  - Consider obtaining data on transactions and positions directly from major market players such as crypto exchanges or crypto asset service providers to reduce reporting burden.
  - Long term: differentiate digital versions of conventional financial assets in separate “of which” positions in respective functional categories (e.g., crypto bonds, crypto derivatives).

*International Monetary Fund — crypto-assets-cg, sections 4.6–4.30 (Brazil, Estonia, Germany).*

### 4.    B.4. GEORGIA

### 4.    B.4. GEORGIA

### Virtual Assets — Mining
- Georgia has emerged as a prominent hub for virtual asset mining, driven by low electricity costs from access to hydroelectric power and the presence of Free Industrial Zones with minimal regulatory burdens.
- Rapid growth in mining activity introduced regulatory and infrastructure challenges.
- Between 2017 and 2021, the surge in electricity demand, primarily linked to virtual asset mining, transformed Georgia from a net exporter to a net importer of electricity.
- Prior to legislative changes, Georgia lacked formal regulations governing virtual asset mining and trading, allowing individuals and entities to operate with minimal oversight.
- Tax treatment prior to reforms:
  - Individual miners and traders were exempt from taxation.
  - Businesses engaged in virtual asset-related activities were generally subject to a 15% corporate income tax and 18% value-added tax.
- Georgia facilitated conversion of virtual assets into fiat directly through bank accounts and allowed both residents and non-residents to purchase real estate using virtual assets.

### Survey Module on Mining
- The National Statistics Office of Georgia (Geostat) has made continuous efforts to identify and survey entities involved in virtual asset-related activities, including analyzing GPU importers and businesses operating in mountainous regions known for mining.
- An ad-hoc module with questions on virtual asset mining activity was incorporated into the broader External Economic Activities of Enterprises survey.
- Despite these efforts, comprehensive data on the sector remain notably lacking.

### Virtual Asset Service Providers (VASPs) and Regulation
- In September 2022, the Parliament of Georgia passed amendments introducing a legislative framework for Virtual Assets (VAs), Virtual Asset Services (VAS) and Virtual Asset Service Providers (VASPs); these amendments became effective on January 1, 2023.
- Amendments covered the Organic Law of Georgia “On the National Bank of Georgia” and the Law of Georgia “On Facilitating the Prevention of Money Laundering and the Financing of Terrorism”.
- Under the legislative framework, the National Bank of Georgia (NBG) has the authority to supervise VASPs, including a mandatory registration regime with requirements on Head office, systems, Fit and Proper Checks, Anti-Money Laundering/Countering the Financing of Terrorism (AML/CFT) Policy Procedures, and AML/CFT supervision.
- VASPs are obliged to comply with the Travel Rule requirement by the end 2026.
- Definition and authorized providers:
  - VASP is an institutional unit that provides virtual asset service for the benefit of another unit.
  - In addition to VASPs, commercial banks, micro-banks, microfinance organizations, and brokerage companies are authorized to provide virtual asset services.
- As of October 2025, there are 28 VASPs registered.

### Institutional scope of virtual asset services (as defined by law)
- Exchange (including via kiosks) between convertible virtual asset and fiat currencies (national or foreign); between one or more forms of virtual assets, between convertible virtual asset and financial instrument — permitted for VASPs, Commercial Bank, Micro-Bank, Microfinance Organisation, Brokerage Company.
- Transfer of convertible virtual asset — permitted for VASPs, Commercial Bank, Micro-Bank, Microfinance Organisation, Brokerage Company.
- Safekeeping and/or administration of convertible virtual asset or of the instrument enabling control over virtual asset — permitted for VASPs, Commercial Bank, Micro-Bank, Microfinance Organisation, Brokerage Company.
- Portfolio management of convertible virtual assets (excluding collective portfolio management) — permitted for VASPs only.
- Administration of the trading platform of the convertible virtual assets — permitted for VASPs only.
- Lending of convertible virtual asset (to legal entities only) — permitted for VASPs only.
- Initial Coin Offering of convertible virtual asset and/or service related to initial coin offering — permitted for VASPs only.

### Challenges and Way Forward
- Mining entities have largely been uncooperative, limiting the effectiveness of data collection and leaving significant gaps in understanding the scale and impact of mining operations.
- High electricity consumption provides indirect insights into virtual asset production but remains an incomplete and imprecise indicator.
- Measuring activities of virtual asset mining companies has proven difficult, contributing to persistent data gaps.
- The National Bank of Georgia is developing a remote supervisory framework (off-site monitoring) requiring VASPs registered in Georgia to report information related to customers, transactions, products, services, counterparties, and delivery channels.
- Expected benefits of the remote supervisory framework:
  - Enhanced supervisory visibility, including over potential cross-border transactions.
  - Improved oversight and alignment with international compliance standards.

*International Monetary Fund.*

### Annex 4.1. Brazil. Sample Foreign Exchange Contract

### Annex 4.1. Brazil. Sample Foreign Exchange Contract

### Foreign Exchange Contract — Template Fields
- Type of foreign exchange contract: ( ) purchase ( ) sale
- Foreign exchange contract number:
- Institution authorized to operate in the foreign exchange market
  - Name:
  - Resident identification number:
  - Address:
  - City, State
- Name of forex settling party:
  - Resident identification number /Foreign ID:
  - Address:
  - City State Country
- Intermediary institution
  - Name:
  - Resident identification number /Foreign ID:
  - Address:
- Transaction data
  - Currency of transaction:
  - Value in foreign currency:
  - Value in domestic currency:
  - FX rate:
  - Contract date:
  - Settlement date:
- Nature of the operation
  - Code:
- Process number:
- Registration number:
- Payment/receipt abroad
  - Correspondent bank abroad:
  - Correspondent bank code abroad:
  - Full name and address of the payer/receiver abroad:
- Other specifications
- Contractual clauses
- Numerical markers appearing in the template: 110, 111

### Glossary of Terms and Definitions Related to Crypto Assets
- Consensus mechanism: Protocol that ensures all participants in a decentralized network agree on the validity of transactions and the current state of the blockchain. Currently, the most important consensus mechanisms in the public blockchains are proof-of-work (PoW) and proof-of-stake (PoS).
- Crypto assets (fungible and non-fungible): Digital representations of value that use cryptography and distributed ledger technology, such as blockchains, to enable parties to transact directly with each other without the need for a trusted intermediary. Two types of crypto assets are distinguished: fungible and non-fungible. Fungible crypto assets are divisible and not unique, and non-fungible crypto assets (most commonly known as non-fungible tokens or NFTs) are unique and non-divisible. Examples of fungible crypto assets include Bitcoin, Ether, and stablecoins such as the USD Coin and Tether, while NFTs include Bored Ape Yacht Club, CryptoPunks, and NBA Top Shot.
- Crypto assets (native and non-native): Native crypto assets originate on their own blockchain and exist only on that ledger. They do not represent, nor derive their value from any assets. Examples: Bitcoin and Ether. Non-native crypto assets represent assets external to the blockchain. These assets function as digital representations of underlying financial or real-world assets, with their value directly linked to those assets. Crypto securities are a notable example of non-native crypto assets.
- Crypto asset transactions (on-chain and off-chain): On-chain transactions are executed and recorded directly on a blockchain’s distributed ledger. They are transparent and immutable but often slower and subject to network transaction fees and commonly associated with DEXs. Off-chain transactions are conducted outside the blockchain. They are not immediately recorded on the distributed ledger, typically allowing faster processing and lower fees, and often used within CEX.
- Crypto assets with a corresponding liability designed to act as a medium of exchange within a platform: Fungible crypto assets that act as a medium of exchange within a platform. They are classified as a separate subcategory under debt securities.
- Crypto assets without a corresponding liability designed to act as a medium of exchange: Fungible crypto assets for which there is no claim on the issuer. They may be designed to act as a general medium of exchange or designed to act as a medium of exchange within a platform only. Example: Bitcoin.
- Crypto custodian: An entity that provides secure storage and management of private keys on behalf of crypto-asset holders, enabling access to and control of their assets. Custodians typically use a combination of hot and cold storage solutions to balance accessibility and security. Examples include Coinbase (which is also a CEX), Fidelity Digital Assets, and Bank of New York Mellon.
- Crypto exchanges (centralized and decentralized): Digital platforms where users can buy and sell various types of crypto assets, including NFTs. They may also offer crypto lending/borrowing facilities, custodial and wallet services. Centralized exchanges (CEX) are platforms that match buyers and sellers of crypto assets through centralized order books and often provide custodial services by holding users’ private keys. Examples: Binance, Coinbase, and Kraken. Decentralized exchanges (DEX) are platforms within the decentralized finance (DeFi) ecosystem that enable users to trade crypto assets directly through smart contracts without intermediaries. They are non-custodial, meaning users retain control of their private keys and transactions are recorded on the blockchain. Examples: Uniswap, SushiSwap, and PancakeSwap.
- Crypto Payment Processors: Service providers that enable businesses to accept payments in crypto assets like Bitcoin, Ether, and stablecoins. They function similarly to traditional payment processors but are tailored for crypto assets. Examples: BitPay and CoinPayments.
- Crypto wallets: Digital tools that allow users to store, manage, and transact with their crypto assets securely. They can be either hot wallets, which are connected to the internet and provide quick access to funds, or cold wallets, which are offline and offer enhanced security for long-term storage.
- Crypto wallet providers: Companies or platforms that offer wallet services. Examples: Ledger, Trezor, and MetaMask.
- Decentralized Finance (DeFi): Range of services within the crypto assets’ ecosystem, aiming to replicate traditional financial services. DeFi business models focus on developing and enhancing investment opportunities in the crypto space. DeFi operates on blockchains like Ether, which support smart contracts.
- Digital assets (designed to act as a medium of exchange or financial instrument): Digital representations of value recorded on a cryptographically secured distributed ledger or using a similar technology or issued by a central bank as CBDC.
- Distributed Ledger Technology (DLT) and Blockchain: Refers to recording data or information (for example on transactions and balances of crypto assets) in multiple places at the same time, ensuring that system/network participants have consistent access to the same data. Blockchain is a type of DLT that has a specific set of features, organizing its data in a chain of blocks. Each block contains data that has been verified, validated, and then “chained”. Blockchains can be categorized into public and private.
- Hot and cold storage of crypto assets: Hot storage of crypto assets refers to wallets that are connected to the internet, making them convenient for quick transactions but more vulnerable to hacks. Cold storage involves offline custody of private keys.
- Means of payment: Refers to the instruments used to make the payment, such as cash, check, debit/credit card, stablecoins, or crypto assets without corresponding liability (e.g., Bitcoins).
- Medium of exchange: A means for acquiring goods, services, or assets without resorting to barter. Acting as a medium of exchange is one of the basic functions of money, which takes the form of various types of financial instruments.
- Miners and validators: Miners are entities that use computational power to solve complex mathematical puzzles in PoW-based blockchains, such as Bitcoin. Once a puzzle is solved, the miner gets the opportunity to verify a set of transactions, add a new block to the blockchain, and receive a reward in the form of newly minted crypto assets. Validators are entities that stake their crypto assets to participate in the transaction verification of PoS-based blockchains, like Ethereum. These entities earn rewards for contributing to the network's security and decentralization.
- NFT issuers: Individuals, organizations, or platforms that create and distribute NFTs, typically representing digital or physical assets such as art, music, collectibles, virtual real estate, or in-game items. These issuers can range from independent artists and musicians to large corporations and gaming companies.
- Private and public keys: Private key is a secret cryptographic key that is used to sign transactions and provide proof of ownership of crypto assets. They allow the owner to access and manage their crypto assets. It grants the holder complete control over the crypto assets associated with the corresponding public key and address. A Bitcoin private key is an alphanumeric string, typically 64 characters long. Public key is a cryptographic key that can be shared with anyone. It is used to encrypt information or to verify a digital signature. The public key can be shared with anyone, allowing them to send encrypted messages or verify transactions signed by the corresponding private key.
- Proof-of-Stake (PoS): Consensus mechanism where validators are chosen based on the number of coins they hold and are willing to lock up "stake" to help validate transactions and maintain the network’s integrity. Examples of crypto assets that use PoS include Ether (after its transition to PoS) and Cardano.
- Proof-of-Work (PoW): Consensus mechanism where participants, known as miners, solve complex mathematical puzzles to validate transactions and add new blocks to the blockchain. This process involves miners competing to solve a cryptographic puzzle, and the first one to solve it gets the opportunity to add the next block to the blockchain, receiving crypto assets as a reward. Prominent examples of crypto assets that use PoW include Bitcoin and Ether (before its transition to PoS).
- Public address: A unique code derived from a public key, used to receive crypto assets. It is in more user-friendly format than public key, shorter than public key, which makes it more practical for sharing. It is the address that users share with others to receive funds in crypto asses such as Bitcoins.
- Pseudonymity: A privacy feature of many public blockchains where users transact with public addresses rather than real-world names, providing partial anonymity. All transactions on a public blockchain like Bitcoin or Ethereum are openly visible, but the participants are identified only by their alphanumeric blockchain addresses and there is no direct way to link theses addresses with real-life identities of holders.
- Security tokens: Crypto assets that represent debt or equity claims on the issuer or derivative contracts that can be exchanged peer-to-peer even if the underlying asset is not a crypto asset.
- Smart contract: Refers to self-executing computer programs stored on a blockchain that automatically enforce and execute predefined rules or transactions once specified conditions are satisfied.
- Stablecoins: A crypto asset that aims to maintain a stable value relative to a specified asset or a pool or basket of assets, usually by being backed by the assets of the issuer or by using algorithms to control supply (so called algorithmic stablecoins—without a claim on the issuer), and sometimes through a combination of both. Examples of i) stablecoins with a claim on the issuer are Tether (USDT) and USD Coin; and ii) algorithmic stablecoins—without a claim on the issuer are Djed, ETHx.
- Stablecoin issuers: Entities other than monetary authorities or government that create, issue, and manage crypto assets with a corresponding liability designed to act as a general medium of exchange.
- Tokenization: Refers to the process of issuing a digital token on a blockchain that represents a financial or tangible asset. In principle, almost any asset can be tokenized—including financial instruments (such as equities or bonds) and physical assets (such as real estate or gold).
- Utility tokens: Fungible crypto assets that provide the holders with future access to goods or services. They are classified as a separate subcategory under debt securities.

---


_Source: https://www.imf.org/-/media/files/data/statistics/bmp7/implementation-support/crypto-assets-cg.pdf_