## hmdtea ## Source details **Canonical URL:** [hmdtea](https://www.imf.org/-/media/files/publications/books/2023/english/hmdtea.pdf) ## Other formats - [Markdown version](/-/media/files/publications/books/2023/english/hmdtea.pdf.md) - [Structured JSON version](/-/media/files/publications/books/2023/english/hmdtea.pdf.json) --- ### Definitions and conceptual framework - Statistical definition: - "Digital trade is all international trade that is digitally ordered and/or digitally delivered." - Digitally ordered trade: "the international sale or purchase of a good or service, conducted over computer networks by methods specifically designed for the purpose of receiving or placing orders." - Digitally delivered trade: "all international trade transactions that are delivered remotely over computer networks" (services only). - Measurement principles: - Digital trade transactions are a subset of existing trade transactions in international merchandise trade statistics and international trade in services statistics. - Accounting follows IMTS 2010, MSITS 2010 and BPM6; updates to those standards will be reflected without changing the statistical definition. - Digitally deliverable services are identified from EBOPS 2010 product lists to provide an upper‑bound; Mode 1 cross‑border supply generally proxies digital delivery. ### Policy importance and motivations - Policy areas affected: - International trade policy (WTO Work Programme on Electronic Commerce; 116 RTAs with digital provisions in 2022 representing 33 per cent of notified agreements). - Low‑value goods and de minimis thresholds (examples: UK GBP 15; US up to US$ 2,000; many countries’ de minimis amounts around 1-3 percent of trade but can exceed 15 percent). - Competition (platform market power), taxation (Pillar One/Pillar Two, VAT on online sales), and development (digital divide concerns). - Specific numeric policy context: - As of July 2023, 89 members participated in the Joint Initiative on Electronic Commerce. - WTO moratorium on customs duties on electronic transmissions extended at the 12th Ministerial Conference in June 2022. ### Key findings and indicators - E‑commerce firm participation (2021 or latest reporting where indicated): - Firms reporting highest share of e‑commerce purchasing: New Zealand 89.6 per cent; Australia 80.7 per cent; Sweden 78.4 per cent; Brazil 75.0 per cent; Canada 74.9 per cent; Netherlands 66.1 per cent. - Firms reporting highest share of e‑commerce sales: Australia 63.3 per cent; New Zealand 60.3 per cent; India 60.2 per cent. - Note: Widespread comparable evidence on domestic vs international split and monetary values is not available. - Growth in digitally deliverable services (global shares of services exports): - 2012: 48 per cent - 2019: 52 per cent - 2021: 63 per cent (pandemic impact) - Selected country case results and statistics: - China CBEC (preliminary aggregates for 2022): total international CBEC trade US$ 309 billion (up 3.8 per cent YoY); CBEC exports US$ 230 billion (up 6.7 per cent YoY); CBEC imports US$ 79 billion (shrank by 3.8 per cent YoY). Consumer goods share of CBEC exports 93 per cent; of CBEC imports 98 per cent. - Costa Rica pilot (2021): 90 per cent of targeted firms digitally delivered services internationally; digitally delivered targeted services amounted to 51 per cent of total services exports and 20 per cent of total exports; contributed 7.2 per cent to GDP in 2021. - BEA 2019 survey (US parents, benchmark): overall share of services digitally ordered 25 per cent; share of goods digitally ordered 19 per cent; foreign affiliates shares between 30 and 35 per cent. - Eurostat household outcome: 21 per cent of individuals (aged 16-74) made online purchases from sellers outside country of residence. - Türkiye ETCD preliminary results: - Monthly average exports value US$ 110 million; monthly average export transactions 1.4 million. - Monthly average imports value US$ 16 million; monthly average import transactions 0.3 million. - Share of ETCD exports (by value) in total exports 0.58 per cent; share of ETCD export transactions in total export transactions 41 per cent. - Share of ETCD imports (by value) in total imports 0.07 per cent; share of ETCD import transactions in total import transactions 21.3 per cent. - Average transaction value: exports US$ 78; imports US$ 55. ### Measurement challenges and data‑source strategy - Principal measurement challenges: - Increased role of small firms and households, rise in low‑value parcels, and prevalence of Digital Intermediation Platforms (DIPs) complicate residency and transaction attribution. - De minimis thresholds often exclude low‑value digitally ordered shipments from customs-based statistics. - Non‑monetary digital flows (data for advertising, free services) are out of scope for digital trade statistics; paid data transactions are in scope when recorded as services. - Recommended data‑source mix: - Combine multiple sources: customs declarations (including ETCDs), low‑value/postal/courier data, business ICT/e‑commerce surveys, core business surveys, ITS surveys, MNE surveys, VAT/OSS/MOSS/OSS tax data, card payments and ITRS. - Integrate household surveys and travel/border surveys to capture household expenditure and Mode 2 consumption. - Use microdata linking where possible (e.g., linking ICT survey responses to trade microdata) and reconcile across sources. ### Digitally ordered trade — operational guidance - Definition and inclusions/exclusions: - Includes orders via web, extranet, EDI, apps, in‑app purchases, online bidding platforms; excludes phone, fax, manually typed email unless country guidance includes email (UNCTAD guidance). - Recurring subscriptions: include all transactions associated with digital orders in the current reporting period; where original ordering channel unknown, estimate share attributable to digital ordering. - Key source roles: - Customs declarations + ETCDs + postal/courier data → merchandise digitally ordered goods (must supplement for low‑value/de minimis). - Business ICT/e‑commerce surveys → monetary values of e‑commerce sales/purchases and breakdowns (domestic vs international, goods vs services, DIPs). - Core business surveys and MNE surveys → broader coverage and large‑firm contributions. - Card payment data → potential proxy for household digitally ordered imports (card‑not‑present transactions, merchant country, MCC codes), subject to access and classification caveats. - VAT/OSS/MOSS data → useful for household imports of digitally delivered services when VAT collected from non‑resident suppliers. - Specific numeric evidences preserved: - EU: around 65 per cent of e‑commerce turnover in EU27 earned by businesses comes from EDI‑type sales. - Canada (2021): 20 per cent of business e‑commerce sales were to customers abroad; over three‑quarters of those orders (by value) went to the United States. - Malaysia survey variable codes preserved (examples): 310002, 310003, 310004, 310011, 310016, 310036, 310043, 310050, 310056–310061. ### Digitally delivered trade — operational guidance and sources - Scope and first step: - Only services considered digitally deliverable/delivered by Handbook convention. - First identify digitally deliverable EBOPS 2010 items (Table 4.1 / Annex C) to compile an upper‑bound of digitally deliverable services. - Two practical approaches to estimate actual digital delivery: - Expert‑judgement shares applied at product disaggregation (e.g., Eurostat‑WTO model; TiSMoS consolidated allocations). - Direct measurement via ITS (International Trade in Services) surveys asking exporters/importers for shares/deliveries by Mode 1/remote delivery. - Examples and model comparisons: - Eurostat‑WTO model shares (2021) vs ITS survey observed shares show variation by product (selected items: accounting 75 model vs exports 51/imports 66; computer services model 75 vs exports 80/imports 56; research and development model 90 vs exports 59/imports 81). - Global 2022 composition of Mode 1 exports of digitally delivered services: Other business services 40 per cent; Computer services 20 per cent; Financial services 16 per cent; Charges for IP n.i.e. 12 per cent; Personal, cultural and recreational services 3 per cent; Insurance and pension services 5 per cent; Telecommunications services 3 per cent; Information Services 1 per cent. - Recommended measurement priorities: - ITS surveys augmented with targeted remote‑delivery questions for digitally deliverable items. - Use business ICT surveys to identify firms delivering digitally and the overlap between digital ordering and delivery. - Use ITRS and VAT OSS/MOSS/OSS data and household ICT surveys to supplement for household consumption and large transactions. ### Digital Intermediation Platforms (DIPs) — accounting and measurement - Definition: - "Online intermediation services that facilitate transactions between multiple buyers and multiple sellers in exchange for a fee, without the online intermediation unit taking economic ownership of the goods or rendering the services that are being sold (intermediated)." - Recording principles: - Distinguish main supply (seller↔buyer) from intermediation services (DIP↔buyer/seller). - Record only intermediation fees for DIPs in the DIP's accounts; do not record the full value of intermediated goods/services as DIP output. - Fees may be explicit or implicit; if payer unknown, assume seller pays the fee. - Digital intermediation services recommended to be recorded under EBOPS 2010 trade‑related services (SJ34). - Measurement approaches: - Survey resident DIPs via ITS, MNE or ICT surveys to collect fees and transacted products. - User‑side (demand) business and household surveys should collect value of goods/services intermediated and, where possible, fees paid and residency of the DIP. - Complement with web scraping, payment card data, app‑store transaction data and targeted surveys of large global DIPs where feasible. - Reporting and balancing example preserved: - Box 5.1 numerical example: buyer pays 100; DIP itemizes 12 fee from buyer and charges 8 to seller; accounting entries yield balanced total imports/exports across Buyer/Seller/DIP countries with net income entries (example numbers preserved in Box 5.1/Table 5.2 context). ### Reporting template and accounting guidance - Reporting template (Table 2.1) structure (items preserved exactly): - Item 1 Total digital trade = 2+3 minus 4 - Item 2 Digitally ordered trade = 2.1+2.2 - 2.1 Goods - 2.1.a of which: via DIPs - 2.2 Services - 2.2.a of which: via DIPs - Item 3 Digitally delivered trade - 3.a of which: via DIPs - Item 4 Digitally ordered and digitally delivered trade - 4.a of which: digital intermediation services - Addenda: - A.1 Digital trade in services = 2.2+3 minus 4 - A.2 Digitally deliverable services = >3 - Annex B (Table B.1) provides 35 illustrative examples mapping transactions (goods/services; digitally ordered/delivered; actor/residence permutations) to reporting template items; compilers may apply flexibility based on available sources and required breakdowns. ### Recommendations and capacity building (selected, preserving numbering and emphasis) - Measurement strategy and prioritization: 1. Assess which institutional sectors (businesses, households, government units, NPISHs) engage in statistically meaningful digital trade and prioritize measurement; prioritize business transactions absent contrary evidence. 2. Use business ICT surveys to measure digitally ordered exports and imports by businesses; collect total e‑commerce sales/purchases and breakdowns (domestic vs international) and product detail (goods, digitally delivered services, other services). 3. Ensure business ICT survey coverage and estimation methods are sufficient to derive representative digitally ordered trade estimates for all businesses; follow international guidelines (UNCTAD Manual; OECD Model Survey). 4. Explore mainstreaming relevant questions into core business surveys used for structural business statistics and/or international trade in services surveys; use MNE surveys and hybrid strategies to combine representative totals with detailed breakdowns. 10. Encourage implementation of the WCO Framework of Standards on e‑commerce and work closely with customs authorities to ensure customs reporting systems identify e‑commerce shipments and reflect statistical needs. 11. Estimate many low‑value international goods shipments (parcel trade), including de minimis, using customs, postal and courier agency data and tax data where coverage adequate. 12. Explore card payment data as a cost‑effective way to estimate household digitally ordered expenditures; address identification and product‑type challenges. 13. Integrate information from different sources to derive digitally ordered trade estimates representative of the entire economy; document sources, coverage, and assumptions to enable comparability. - Digitally delivered trade recommendation highlights: - Target digitally deliverable EBOPS items (Table 4.1 / Annex C) first. - Use ITS surveys to collect country‑specific Mode 1 shares or adopt the Eurostat‑WTO expert judgment model where surveys are not available, and regularly reassess shares for country relevance. - Combine ITS survey measurements with ICT/business surveys, ITRS and VAT/OSS data to capture household and large enterprise transactions. - DIPs recommendations: - Distinguish intermediation fees from intermediated transaction values in surveys and administrative sources. - Include digital intermediation service questions in ITS and enterprise ICT surveys; collect fee payer information and residency. - Consider targeted engagement and data‑sharing with large global DIPs under data‑sharing agreements where legal frameworks permit. ### Country case practical lessons (selected) - China Customs CBEC implementation: - Uses dedicated customs procedure codes (examples: 9610, 1210, 1239, 9710, 9810), enterprise surveys, platform financial reports, parcel sampling and big‑data methods to estimate CBEC. - 2017 CBEC estimate US$ 13.3 billion (based on subset codes) and 2019 US$ 27.5 billion (codes combined); 2022 total CBEC US$ 309 billion. - Methodology examples preserved: import compilation Vi = Vi1 + Vi2 / Rsi; export compilation Ve = (Ve1 + Ve2) / Rse with Ve1 domestic platform exports and Ve2 overseas platform estimates. - MOFCOM service outsourcing monitoring: - Online Monitoring System on Service Outsourcing (since 2009) and Investigation System underpin ITO/BPO/KPO compilation; international service outsourcing 2022 value US$ 136.8 billion (2019 value US$ 96.9 billion). - International ITO in 2022 US$ 56.1 billion (2019 US$ 42.7 billion). - Jamaica stocktaking: - Identified use of Survey of Living Conditions, tourist surveys, credit card data (Bank of Jamaica), customs data, TAJ VAT returns and quarterly enterprise survey as potential inputs; noted major limitations (monetary values, product detail, domestic vs cross‑border differentiation). - Türkiye pilot using ETCDs and customs declarations: - ETCD preliminary monthly averages: exports value US$ 110 million with 1.4 million transactions; imports value US$ 16 million with 0.3 million transactions; average transaction values exports US$ 78 and imports US$ 55. - Challenges: EDI detection, identification of DIPs, distinguishing B2B/B2C/C2C and product code granularity; quality assurance, guidance and training implemented. ### Data transparency and communication - Compilers should: - Record and communicate sources, coverage, estimation methods, and limitations (including firm size, industry coverage, institutional sectors). - Report overlaps between digitally ordered and digitally delivered trade (item 4 in reporting template) and subtraction rules to avoid double counting. - Use standardized product classifications (EBOPS 2010 for services; HS/CPC for goods) and maintain metadata on methodological choices. *Source: Handbook on Measuring Digital Trade — IMF, OECD, UNCTAD and WTO (Foreword, Executive summary, Chapters 1–6, Annexes and Boxes as provided in the source content).* ### Foreword ### Foreword ### Acknowledgements and Contributors - This second edition of the Handbook on Measuring Digital Trade is the outcome of a collaborative effort by the International Monetary Fund (IMF), the Organisation for Economic Co-operation and Development (OECD), the United Nations Conference on Trade and Development (UNCTAD) and the World Trade Organization (WTO). - Authors: Patrick Quill (IMF); Antonella Liberatore, David Brackfield and Julia Schmidt (OECD); Daniel Ker (UNCTAD); Barbara D’Andrea Adrian, Joscelyn Magdeleine and Ying Yan (WTO). - Management: Carlos Sánchez Muñoz, Maria Borga and Kristy Howell (IMF); Annabelle Mourougane (OECD); Torbjörn Fredriksson (UNCTAD); Barbara D’Andrea Adrian (WTO). - Case study contributors (selected): Xiaoyuan Zhai, Qian Li, Zheng Fang and Weiguo Qi (General Administration of China Customs); Xuyang Wang, Yanhui Jing and Yizhen Xie (China’s Ministry of Commerce (MOFCOM)); Esmond McLean (Bank of Jamaica); Hank Williams and Diedre Campbell (Tax Administration Jamaica (TAJ)); José Antonio Isanta Foncuberta (National Statistics Institute of Spain (INE)); Çaglayan Aslan (Türkiye Ministry of Trade); Aylin Kolbası, Eyüp Mehmet Dìnç and Esengül Tanrikulu (Turkish Statistical Institute (TurkStat)). - Additional substantive inputs: Jennifer Bruner and Alexis Grimm (United States Bureau of Economic Analysis (BEA)); John Mitchell (OECD). - Important contributions from IMF, OECD, and WTO staff and from numerous national statistical offices and central banks across Argentina, Australia, Brazil, Canada, Costa Rica, Denmark, Germany, Hungary, Indonesia, Ireland, Israel, Italy, Kenya, Mexico, Netherlands, United Kingdom, United States, Eurostat, and others. - Production management and editing: Anthony Martin and Serge Marin-Pache (WTO); edited by Helen Swain (WTO). - Acknowledgement that contributors to the first edition are also thanked. ### Foreword: Purpose and scope - Digital technologies increasingly enable global placement and receipt of orders, and instantaneous remote delivery of services into businesses and homes, including internationally. - The Handbook focuses on two criteria—digital ordering and digital delivery across borders—to offer a conceptual and measurement framework for digital trade aligned with broader macroeconomic statistics. - This second edition: - Builds upon the first edition (2019) and benefits from substantive inputs by both developed and developing economies. - Leaves the fundamental measurement framework unchanged, while providing clarifications to concepts and definitions related to digital trade and guidelines for operationalization. - Reflects advances statistical compilers have made in measuring digital trade and expands compilation guidance based on national and international efforts. - Proposes a revised reporting template offering flexibility for compilers when collating components of digital trade, including when only partial information is available. - The Handbook establishes a shared foundation for understanding and measuring digital trade in an internationally comparable way and supports technical assistance and statistical capacity-building programs by the four co-authoring organizations. ### Executive summary — Key findings and recommendations - Digitalization is transforming how products are purchased and delivered; yet traditional macroeconomic statistics often leave digitalization largely invisible because they focus on what is produced and who produces it. - The Handbook’s statistical definition (from the first edition) formalizes digital trade as: “digital trade is all international trade that is digitally ordered and/or digitally delivered”. - The statistical definition: - Reflects the multidimensional character of digital trade by identifying the nature of the transaction as the defining characteristic. - Is the basic building block of a conceptual measurement framework consistent with broader macroeconomic accounts. - Measurement principles: - Digital trade transactions are a subset of existing trade transactions measured in international merchandise trade statistics and in international trade in services statistics. - Any economic actor can engage in digital trade; accounting principles follow International Merchandise Trade Statistics Concepts and Definitions (UN, 2011), the Manual on Statistics of International Trade in Services (UN et al., 2010), and the Balance of Payments (IMF, 2009). - The Handbook’s framework aligns with broader macroeconomic standards; updates to those standards (notably any change in the production boundary) will be reflected in the measurement framework without changing the statistical definition of digital trade. - Concepts are in line with guidance on measuring the digital economy established through the framework for digital supply and use tables (OECD, 2023). - Measurement challenges and data-source strategy: - Digital ordering and delivery exacerbate known measurement challenges: increased involvement of small firms and households in international trade; rise in low-value trade; and the growing role of digital intermediation platforms (DIPs) adding a third actor to transactions. - Traditional data sources—often reliant on large firms—may inadequately cover these developments. - Recommendation: build on and combine existing data sources to produce comprehensive digital trade statistics; discuss benefits and limitations of each source; integrate case studies and national experiences. - Digitally ordered trade (international e-commerce): - Defined as “the international sale or purchase of a good or service, conducted over computer networks by methods specifically designed for the purpose of receiving or placing orders”. - Covers transactions in both goods and services; business transactions account for the bulk of e-commerce sales and purchases (UNCTAD, 2023). - ICT surveys and enhanced business surveys are recommended to collect monetary values of e-commerce transactions and to break out digitally ordered trade. - Household and/or travel surveys can measure e-commerce transactions by individuals. - Administrative sources and card payment data can measure components of digitally ordered exports and imports; customs records are particularly relevant if digitally ordered shipments are flagged with specific customs procedure codes. - Accurate estimation of low-value trade is necessary for exhaustive coverage; sources include customs administrative data and postal and courier agency information. - No single source provides a holistic measure; integrate multiple sources to cover all institutional units in the whole economy. - Digitally delivered trade: - Defined as “all international trade transactions that are delivered remotely over computer networks”; the Handbook takes the view that only services can be digitally delivered. - Digital delivery can occur over a longer period and can involve remote inter-personal interaction through computer networks. - First step: identify service items that are digitally deliverable (i.e., can be delivered through computer networks). - Aggregating identified items from existing statistics provides an upper-bound estimate of digitally delivered trade without changes to data collection mechanisms. - For digitally deliverable services, cross-border supply (Mode 1) can be considered equivalent to digital delivery; trade in services by mode of supply can provide reasonable estimates for digitally delivered trade. - Most countries are only beginning to measure trade in services by mode of supply; where such data are absent, shares based on expert judgement (for example, the Eurostat-WTO model (Eurostat, 2021a and WTO, 2023)) may be used but should be regularly assessed for country-specific relevance. *Handbook on Measuring Digital Trade — Foreword and Executive summary* ### Executive summary ### Executive summary ### Measurement approaches and data sources - International trade in services (ITS) surveys, which cover businesses, provide the best means for obtaining direct estimates of digitally delivered services trade. - ITS surveys can be enhanced with supplemental questions (for example the model developed by UNCTAD in collaboration with Costa Rica, India, and Thailand (UNCTAD, 2021a)) so that shares of digitally delivered exported and imported services can be measured in an integrated way with overall services trade sources and methods. - Firm-based sources are likely to cover the bulk of digitally delivered trade; however, households are increasingly involved in digitally delivered services trade, so: - Statistical compilers must investigate how household surveys and other data sources can be used to improve coverage of digitally delivered trade estimates. - Some digitally delivered services may be consumed while abroad (i.e., supplied via Mode 2) and would therefore require different estimation strategies, even if the values are often not economically significant. - Information from various sources should be integrated so that digitally delivered trade estimates representative of the entire economy can be derived. - Surveys of ICT usage in business are better placed to collect information on the transacted products (i.e., the goods and services being intermediated), whereas ITS surveys are recommended to collect information on the exports and imports of intermediation services by businesses. - To ensure household-sector coverage, several types of household surveys should include questions on: - the value of goods and services purchased via DIPs (notably from well-known DIPs and for tourism-related services), and - ideally, on the intermediation fees paid. ### Digital intermediation platforms (DIPs) - Definition: DIPs are “online interfaces that facilitate, for a fee, the direct interaction between multiple buyers and multiple sellers, without the platform taking economic ownership of the goods or rendering the services that are being sold (intermediated)”. - Rationale for special treatment: Although transactions intermediated by DIPs are, in principle, included in existing trade statistics and covered by digitally ordered and digitally delivered trade concepts, DIPs are highlighted because of their significant role in the economy, policy interest, and specific compilation challenges. - Nature of the service: The service provided by DIPs is “matching” buyers with sellers, facilitating exchange of goods or provision of services. These digital intermediation services are, by definition, both digitally ordered and digitally delivered and are defined as “online intermediation services that facilitate transactions between multiple buyers and multiple sellers in exchange for a fee, without the online intermediation unit taking economic ownership of the goods or rendering the services that are being sold (intermediated)”. - Recording guidance: - Distinguish the supply of goods or services (transaction between seller and buyer) from provision of intermediation services (transaction between the DIP and both the seller and the buyer). - Regardless of whether a DIP facilitates transactions in goods or services, intermediation fees should be recorded under trade-related services in the international accounts. - When a DIP is resident in the compiling economy, surveys can be used to measure both exports of intermediation services and the underlying goods and services transactions. - Data collection recommendation: Explore several data sources and collect information on exports and imports of intermediation services via ITS surveys; complement with ICT usage and household surveys to capture transacted products and household purchases. ### Reporting digital trade transactions - The Handbook proposes a reporting template supporting compilation of: - digitally ordered trade, - digitally delivered trade, and - total digital trade. - The template allows different components to be measured in the way that best suits the compiler, even when only partial information is available. - For a comprehensive measure of total digital trade, it is important to develop data sources that can measure digitally ordered trade, digitally delivered trade, and also identify trade that is both digitally delivered and digitally ordered. - ICT usage surveys (for both businesses and households) are well placed to measure the overlap between digitally ordered and digitally delivered trade. - Surveys should collect information on sales and purchases broken down by: - goods, - digitally delivered services, and - other services. ### Key findings and indicators - E‑commerce firm participation (2021, latest reporting where indicated): - Firms reporting the highest share of e-commerce purchasing: New Zealand (89.6 per cent), Australia (80.7 per cent), Sweden (78.4 per cent), Brazil (75.0 per cent), Canada (74.9 per cent), Netherlands (66.1 per cent). - Firms reporting the highest share of e-commerce sales: Australia (63.3 per cent), New Zealand (60.3 per cent), India (60.2 per cent). - Note: Widespread and comparable evidence on the split between domestic and international e-commerce, as well as on the value of e-commerce transactions, is not available. - Growth in digitally deliverable services (global): - In 2012, digitally deliverable services represented 48 per cent of global exports of services. - In 2019, the share increased to 52 per cent of global exports of services. - In 2021, the share jumped to 63 per cent of global exports of services, reflecting the impact of the COVID-19 pandemic on composition and nature of internationally traded services. - Historical initiatives and definitions: - In 2012, the United States Bureau of Economic Analysis (BEA) attempted to measure trade in “digitally enabled” services. - In 2013, the Task Group on Measuring Trade in ICT Services and ICT-enabled Services (TGServ), led by UNCTAD, developed recommendations and indicators on ICT-enabled services trade; this Handbook builds on that work to define digitally deliverable services. - Platforms and marketplaces: - Online platforms act as market makers and have transformed retail, wholesale, accommodation, transport, food services, and many B2B transactions; sales of goods and services through online platforms surged during the COVID-19 pandemic. ### Statistical definition and conceptual framework - Adopted statistical definition (from first edition and refined): - “Digital trade is all international trade that is digitally ordered and/or digitally delivered.” - Digitally ordered trade: “the international sale or purchase of a good or service, conducted over computer networks by methods specifically designed for the purpose of receiving or placing orders.” - Digitally delivered trade: “all international trade transactions that are delivered remotely over computer networks” (covers services only and is broader than earlier wording, as it covers any form of digital delivery). - Rationale: - The two criteria (digital ordering and digital delivery) identify the nature of the transaction as the defining characteristic and acknowledge potential overlap. - Alignment with OECD, UNCTAD, WTO initiatives and with e-commerce and ICT-enabled services measurement efforts provides clarity and allows compilers to leverage existing measurement instruments. - The concepts of digital ordering and digital delivery have been integrated into the framework of digital supply and use tables. ### Moving forward and capacity building - The Handbook provides a consistent measurement framework to guide compilers in measuring digital trade. - Further research and empirical testing will be needed to improve and refine compilation approaches, but the fundamental conceptual framework is well established and constitutes the basis for internationally comparable statistics on digital trade. - The Handbook provides the foundation for an active programme of technical assistance and statistical capacity-building by the four partner organizations: the International Monetary Fund (IMF), the Organisation for Economic Co-operation and Development (OECD), the United Nations Conference on Trade and Development (UNCTAD) and the World Trade Organization (WTO), to support statistical compilers as they measure, monitor and respond to the challenges of digital trade. *Source: Executive summary, Handbook on Measuring Digital Trade* ### 1.4 Measuring digital ### 1.4 Measuring digital trade is key for effective policymaking ### Policy importance and motivation - The Handbook aims to help statistical compilers address policymakers’ demands for better statistical evidence on digital trade. - Since the G20 Chinese presidency in 2016, the G20 has emphasized measurement of the digital economy and digital trade. The 2017 G20 Digital Economy Ministerial Declaration called for a review of statistical frameworks to separately identify the digital economy. - Measurement is crucial to enable policymakers to harness, regulate and shape digital trade flows and to assess market access in a rapidly changing business environment. - Digital trade interacts with many domestic and international policy areas, including competition, tax, development and economic growth. ### 1.4.1 International trade policy - Digital trade policy discussions occur in: - WTO Work Programme on Electronic Commerce (launched 1998). - WTO Joint Initiative on Electronic Commerce. - Regional trade agreements (RTAs) and digital economy agreements (DEAs). - Evidence constraints: discussions are taking place "in the context of a relatively thin evidence base," limiting understanding of benefits, transmission channels and policy implications. - Multilateral agreements relevant to digital trade: - General Agreement on Trade in Services (GATS) — entered into force January 1995. - General Agreement on Tariffs and Trade (GATT) — entered into force October 1947. - Trade Facilitation Agreement (TFA) — entered into force February 2017. - Information Technology Agreement (ITA) — concluded December 1996. - WTO moratorium on customs duties on electronic transmissions: agreed in 1998; most recently extended at the 12th Ministerial Conference in June 2022. - As of July 2023, 89 members were participating in the Joint Initiative on Electronic Commerce addressing issues including e-signatures, e-payments, information flows, privacy, consumer protection and cybersecurity. - RTAs and DEAs: - In 2022 there were 116 RTAs with digital trade provisions, representing 33 per cent of all agreements notified to WTO. - Examples of DEAs: Digital Economy Partnership Agreement (DEPA) between Chile, New Zealand and Singapore; DEA between Australia and Singapore. - DEAs extend discussions to areas such as artificial intelligence (AI). - Domestic regulation affecting digital trade has become increasingly tight according to the OECD Digital Services Trade Restrictiveness Index (DSTRI); infrastructure and connectivity measures (including restrictions on cross-border data flows and data localization requirements) account for the bulk of the increase in the index. - Policy need: quantifying digital trade flows accurately and comparably would support trade policy discussions and enable analysis of the implications of national regulatory changes. ### 1.4.2 Trade in low-value goods and de minimis thresholds - Digitization facilitates trade through easier data exchange, faster customs clearance and improved risk management. - WCO “Framework of Standards on cross-border e-commerce” requires legal and regulatory frameworks for advance electronic data exchange and enhanced facilitation and control measures. - Emergence of online platforms has increased cross-border flows of low-value packages, creating opportunities for individuals and MSMEs and new measurement and policy challenges. - De minimis thresholds can exempt goods from customs duties and/or taxes and entitle them to expedited procedures with fewer documentation requirements; values below these thresholds are often not recorded. - UN recommendation: estimate low-value trade in merchandise trade statistics (UN, 2011). - Handbook contribution: guidance on improving official measures of digitally ordered goods (see Chapter 3) to better capture low-value trade. ### 1.4.3 Competition policy - Online platforms can deliver consumer benefits (lower prices, greater accessibility, more variety) but also raise concerns about: - Anti-competitive conduct. - Durable market power, reduced market dynamism and innovation. - Systemic risks and rent-seeking behavior. - Importance of identifying large digital intermediaries separately within digital trade statistics to assess their role in international trade. ### 1.4.4 Taxation policy - Digitalization allows firms to export to markets without a physical presence, challenging taxing rights based on physical presence and exacerbating profit-shifting by MNEs relocating intangible assets to low-tax jurisdictions. - Relevance to BEPS two-pillar solution (OECD/G20): - Pillar One: reallocating taxing rights among economies for the share of profits of the world’s largest and most profitable MNEs; reallocating some taxing rights to markets where MNEs have business activities and earn profits regardless of physical presence. - Pillar Two: ensuring all MNE groups with an annual turnover of more than EUR 750 million will be subject to a minimum effective tax rate of 15 per cent. - Indirect taxation challenges: digitalization has created challenges for VAT imposition and collection on online sales of services and physical goods in international trade. - For many countries VAT is the single largest source of indirect tax revenues; in several developing economies VAT is the single largest source of all tax revenues (OECD, 2022). - Accurate measurement of digital trade will help governments adapt taxation frameworks; VAT data can be a useful source to measure elements of digitally ordered and digitally delivered trade (see Chapters 3 and 4). ### 1.4.5 Economic growth and development - Digital trade offers significant opportunities, but many developing economies lag in infrastructure, skills and regulatory environments needed to benefit. - Market openness is necessary but not sufficient; comprehensive policy action across skills, trade, competition, taxation, innovation and connectivity is required to avoid a growing digital divide. - At the 12th WTO Ministerial Conference, members agreed to reinvigorate work under the Work Programme on Electronic Commerce with attention to development dimensions. - WTO members agreed that digital connectivity will be one of three priority areas in the WTO Aid for Trade work programme for 2023-24. - Development cooperation should help bridge gaps in digital connectivity and IT to support business enabling environments in developing countries and LDCs. - Handbook highlights developing economy experiences producing insights on digital trade (see Chapter 6). ### Purpose and structure of the Handbook (Section 1.5) - Objective: provide compilers with a statistical definition of digital trade, a conceptual measurement framework and practical compilation guidance to make digital trade transactions more visible in international merchandise and services trade statistics. - This second edition builds on the first edition (OECD, WTO and IMF, 2019), providing conceptual clarifications while keeping the definition and measurement framework broadly unchanged. - Contributions of the Handbook: - Providing a statistical definition of digital trade and its components; - Establishing a conceptual framework on how to measure digital trade; - Proposing a reporting template to record digital trade transactions; - Providing specific compilation guidance; - Sharing best practices and case studies. - Chapter organization: - Chapter 2: conceptual framework and reporting template for digital trade. - Chapter 3: measurement of digitally ordered trade. - Chapter 4: measurement of digitally delivered trade. - Chapter 5: recording transactions enabled by digital intermediation platforms. - Chapter 6: detailed case studies contributed by China, Jamaica, Spain and Türkiye. - The Handbook builds on existing compilation practices and contributions from national compilers, while recognizing that coordinated international effort is still required to address remaining practical and conceptual challenges. *Source: Handbook on Measuring Digital Trade, IMF, OECD, UNCTAD and WTO, 2023.* ### 1.6 Areas of ongoing work ### 1.6 Areas of ongoing work ### 1.6.1 DIGITALIZATION, INVESTMENT AND INTELLECTUAL PROPERTY - Digitalization exacerbates measurement challenges for international trade, notably for intellectual property products that can be “moved” across international borders within the same MNE, complicating attribution of economic ownership and determining the type and direction of related transactions (IMF, 2008). - Digitalization blurs lines between cross-border services transactions (balance of payments) and services sales/output through the establishment of foreign affiliates; this is especially acute for digital intermediation platforms and platforms providing access to intellectual property product content (e.g., streaming platforms). - While these complexities do not undermine the Handbook’s conceptual framework, they call for further guidance on feasible and comparable compilation approaches based on country experiences. ### 1.6.2 CROSS‑BORDER DATA FLOWS - Cross-border data flows are increasingly central to trade and production and raise policy concerns (privacy protection, digital security, national security, regulatory reach, competition, industrial policy). - Better measurement is needed of: - the volume of international data flows, and - the conditions under which data cross borders effectively. - Some international data flows are direct manifestations of digital trade (arising when an order is placed or a service is delivered through computer networks); their economic value is accounted for by recording the value of the transaction they facilitate in digital trade. - Data assets traded internationally as products (e.g., databases) are accounted for in digital trade when they take the form of monetary transactions in services. - Not all cross-border data flows arise from or relate to trade transactions (UNCTAD, 2021b). - Measuring and recording the value of assets based on the data underlying goods and services transactions is under discussion in the update of macroeconomic statistical frameworks; understanding implications requires further research and experimentation. ### 1.6.3 CLOUD COMPUTING - Cloud computing services are defined as “computing, data storage, software, and related IT services accessed remotely over a network, supplied on demand and with measured resource usage that allows charging on a pay-per-use basis”. - Cloud computing is increasingly replacing ownership of on‑premises IT equipment. - Main suppliers are MNEs with operations spanning many countries and a potentially global customer base. - Paid international transactions in cloud computing should be recorded as trade in services, but determining where the service originated and where it was consumed is challenging in practice, even when the payer and payee countries are known. - Cloud computing often relies on international data transfers between related parties that take place without a corresponding monetary transaction (IMF, 2022). ### 1.6.4 NON‑LIABILITY CRYPTO ASSETS - Crypto assets are defined as “digital representations of value that rely on cryptography and decentralized peer-to-peer architecture based on distributed ledger technology (DLT), which enables two parties to directly transact with each other without the need for a trusted intermediary”. - Non-liability crypto assets (designed to act as a general medium of exchange without a corresponding liability, such as Bitcoin and Ether) were agreed in March 2023 to be treated as non-produced non-financial assets in the updates to BPM6 and SNA 2008, and therefore excluded from the scope of digital trade. - The recommendation can be revisited if there are significant market, regulatory and/or accounting changes, either before or after the release of the manuals in 2025. - The Handbook’s measurement framework is generally consistent with BPM6; consequently, no crypto asset is currently considered in scope for measuring digital trade. - Note: Fungible crypto assets with corresponding liability, such as stable coins with a claim on the issuer, are considered as financial assets and are also not in scope for international trade. ### 1.6.5 COMPLEMENTARY GUIDANCE ON MEASURING THE VALUE OF E‑COMMERCE - Digitally ordered trade is a subset of total e-commerce, specifically where seller and buyer are in different economic territories. - At its November 2022 meeting, the UNCTAD Working Group on Measuring E-commerce and the Digital Economy established a task team to discuss international measurement practices and to work toward developing guidelines to support countries in measuring the value of e-commerce. - Given the relationship between e-commerce and digitally ordered trade, this effort will support operationalization and adoption of the Handbook’s compilation guidance (Chapter 3) on using surveys to measure digitally ordered trade. *Handbook on Measuring Digital Trade — Section 1.6 Areas of ongoing work* ### 2. Digitally ordered transactions can involve ### 2. Digitally ordered transactions can involve ### Digitally ordered transactions (definitions and clarifications) - Digitally ordered transactions can involve participants from all institutional sectors (shown in the “Actors” column of Figure 2.1). - Digitally ordered transactions cover orders made over the web, extranet or via electronic data interchange (EDI, see Box 2.1). - Digitally ordered trade includes purchases of applications (apps) and in-app online purchases. - Digitally ordered trade includes transactions via online bidding platforms. - Orders made by phone, fax or manually typed email are excluded from digitally ordered trade. - Offline transactions formalized using digital signatures are excluded from digitally ordered trade. - Each trade transaction should be treated separately. When a transaction is established via offline ordering processes, but subsequent transactions (or follow up orders) are made via digital ordering systems, the follow-up orders should be considered as e-commerce. - Trade transactions do not necessarily coincide with contracts. For a contract spanning several statistical periods and potentially involving multiple transactions, each transaction should be classified as digitally ordered or not digitally ordered, reflecting the mode(s) of ordering initiated in the current period. - Areas of ambiguity remain and are subject to further research (for example, purchases via online chat functions such as WeChat or WhatsApp may be ambiguous: chat functions are typically not specifically designed for placing orders and receive manually composed messages similar to emails, which are excluded; however, if workflows are automatized using artificial intelligence (AI) such that manually typed orders are processed automatically, the related transactions could arguably be classified as digitally ordered trade). ### Digitally delivered transactions (definition and clarifications) - Definition: “All international trade transactions that are delivered remotely over computer networks.” - This definition is broader than the previous Handbook version and avoids needing to interpret “specifically designed” delivery methods. - The revised definition aligns with the concepts of ICT-enabled services and of cross-border supply of services (Mode 1, see MSITS 2010 (UN et al., 2010) and Box 2.2). - Digitally delivered trade is, de facto, equivalent to “ICT-enabled services” as defined in UNCTAD (2015): “services products delivered remotely over ICT networks.” - Digitally delivered trade only covers services; goods cannot be digitally delivered in this Handbook. - Some Mode 1 services (e.g., most transport services and postal delivery) are not digitally deliverable; remaining Mode 1 estimates can be considered digitally delivered trade. - Services supplied via presence of natural persons (Mode 4) cannot be digitally delivered. Supporting clarifications to identify digitally delivered transactions: 1. Only services can be digitally delivered; 2. Digitally delivered transactions can involve participants from all institutional sectors; 3. For digitally delivered transactions, the payment for and ordering of the services do not have to be conducted online; 4. Services delivered by phone, fax, video call or email are included in digitally delivered trade; 5. Digitally delivered trade includes services provided through apps; 6. Each trade transaction should be treated separately. When a trade transaction is delivered via offline processes, but subsequent follow-up transactions are delivered digitally, the follow-up transactions should be considered as digitally delivered; 7. A trade transaction can be delivered via multiple (digital and non-digital) modes. - Transactions in products such as most insurance services (notably, the core service of risk management) and financial services (such as liquidity provision and transformation, underwriting, safekeeping, record-keeping and payment services) are assumed to be in scope for digitally delivered trade due to the enabling role of computer networks (see Chapter 4). Box 2.1 (key points on computer networks and EDI) - “The internet is a worldwide public computer network”. - “Other computer networks include internal networks (e.g., a LAN), proprietary external networks which are not IP‑based (for instance, the networks set up for early versions of EDI), and automated telephone systems”. - Electronic data interchange (EDI) is the computer‑to‑computer transmission of business data in an electronic format using agreed standards; messages are composed and processed without human intervention. - Practically, computer networks are equivalent to the concept of “ICT networks” (voice or data networks, including the internet). The devices used to access those networks and precise network features do not affect whether a transaction is considered digitally ordered or delivered. ### Transactions enabled by Digital Intermediation Platforms (DIPs) - OECD definition of online platforms (2019): “a digital service that facilitates interactions between two or more distinct but interdependent sets of users (whether firms or individuals) who interact through the service via the internet.” - Key defining features of DIPs (from WCO (2022) and OECD): 1. There are multiple buyers and multiple sellers that interact through the platform; and 2. The platform itself does not own the goods, nor does it render the services that are being intermediated. - Handbook definition of DIPs: “Online interfaces that facilitate, for a fee, the direct interaction between multiple buyers and multiple sellers, without the platform taking economic ownership of the goods or rendering the services that are being sold (intermediated).” - The Handbook assumes all transactions undertaken via a DIP are digitally ordered. Often the products advertised can only be paid for electronically (though means of payment do not affect digitally ordered/delivered classification). - Services offered by platforms that intermediate electronic content without taking economic ownership of the intellectual property products they distribute (such as app stores) are included in DIPs; a DIP is deemed not to take economic ownership if the licence-holder of the intellectual property does not charge the online platform for distributing the digital content until after the consumer has paid. Rationale for separately highlighting DIPs: 1. Interest in the economic role of DIPs and their potentially transformative impact on the economy; 2. A targeted focus on DIPs may deliver (partial) results on both digitally ordered and digitally delivered trade through dedicated survey vehicles; 3. Specific conceptual and statistical challenges when the DIP is not resident in the economy where the intermediation services are consumed (see Chapter 5). - When identifying international transactions undertaken via DIPs, record both the value of the transaction between buyer and seller and the fee accruing to the DIP. - Definition of digital intermediation services in this Handbook: “Online intermediation services that facilitate transactions between multiple buyers and multiple sellers in exchange for a fee, without the online intermediation unit taking economic ownership of the goods or rendering the services that are being sold (intermediated).” - DIPs are remunerated via fees from buyer, seller, or both. Fees can be separately itemized, implied by price differences, collected at the same time as the main transaction, or collected separately (e.g., monthly subscription). Fees accrue to the DIP rather than to the seller. - Fees for digital intermediation services should be separately measured or estimated (see Chapter 5). ### The product dimension: Goods and Services - The framework splits products into goods and services (see Figure 2.1). - Goods: - The Handbook adopts the convention that goods cannot be delivered digitally. - Therefore, goods trade relevant for measures of digital trade comprises only those goods that have been digitally ordered. Any good can be digitally ordered. - Services: - Digital trade in services is broken into digitally ordered services and digitally delivered services; these components overlap (including digital intermediation services). - Digitally ordered services: Transactions in services that are digitally ordered should be included as digitally ordered services (this includes digitally ordered services that are not digitally delivered and services that are both digitally ordered and delivered). - Digitally delivered services: Builds on ICT-enabled services work from the UNCTAD-led Task Group (TGServ). The Task Force identified CPC Version 2.1 products which can potentially be ICT-enabled; this list underpins the Handbook’s list of services considered “digitally deliverable” (see Chapter 4 and UNCTAD, 2015). ### Additional institutional context - The OECD Informal Advisory Group on Measuring GDP in a Digitalised Economy (created in 2017) focuses on improving measurement of digitalization in national accounts, including better visibility of data, AI, DIPs and free digital services. - The Advisory Group developed the Digital Supply and Use Tables (Digital SUTs) framework and is overseeing creation of a handbook on compiling Digital SUTs, a companion to this Handbook. *Source: IMF, OECD, UNCTAD and WTO.* ### CHAPTER 2 — THE CONCEPTUAL FRAMEWORK FOR MEASURING DIGITAL TRADE ### CHAPTER 2 — THE CONCEPTUAL FRAMEWORK FOR MEASURING DIGITAL TRADE ### Digital intermediation services - Digital intermediation services provided by DIPs to the buyers and sellers whose transactions the DIP intermediates are recorded in digitally ordered and digitally delivered services trade when the DIP is resident in a different economy to the buyer/seller (including if the buyer and seller are resident in the same economy as one another). - The Handbook recommends recording these services in the Extended Balance of Payments Services Classification 2010 (EBOPS 2010) under trade-related services (SJ34), a subcomponent of other business services (SJ). ### Actors (Who) - Any economic actor can engage in digital trade; separate identification of different actors can provide important policy-relevant insights. The recommended reporting template does not incorporate a breakdown by actor, but compilers are encouraged to explore relevant breakdowns. - Corporations - Corporations undertake the majority of international trade and can be expected to account for the bulk of digital export and import flows. - Besides DIPs, other important online operators include: - E‑tailers: - Defined as “retail and wholesale businesses engaged in purchasing and reselling goods, which receive a majority of their orders digitally” (OECD, 2023). - E-tailers own the products sold and provide margin based distribution services, as opposed to digital intermediation services. - DIP and e-tailing business models may co-exist within the same enterprise; measurement efforts should focus on the nature of individual transactions facilitated by hybrid platforms. - Online transactions via e-tailers are digitally ordered but do not entail the provision of digital intermediation services. - Other producers only operating digitally: - Businesses that produce their own services for sale but operate exclusively digitally (e.g., priced digital media providers, subscription-based digitally delivered services). - Streaming platforms, cable television and radio subscription services are included because they are deemed to assume economic ownership of the intellectual property products they distribute before streaming. - Transactions via such producers are digitally ordered and digitally delivered but do not involve digital intermediation services. - Data‑ and advertising‑driven digital platforms: - Platforms operating exclusively online, facilitating non-monetary interactions, providing services free to end-users and generating revenue by selling data or advertising (examples: free social media, dating apps, search engines, knowledge-sharing platforms). - Includes websites/platforms receiving revenue for directing visitors to third-party websites; such processes do not explicitly facilitate transactions between two independent sets of users. - Interactions facilitated by these platforms are generally not in scope for measures of digital trade, but compilers should be aware of blurred business-model lines and consider the nature of individual transactions. - Some platforms may charge fees and thus facilitate transactions similar to DIPs (example: Facebook Marketplace charging “selling fees”). - Additional platform categories: - Fee-based platforms facilitating peer-to-peer lending, equity-based crowdfunding, philanthropic crowdfunding, fee-based interest-group platforms, and fee-based dating apps; these services are generally digitally ordered and digitally delivered. - Households - Technological change has increased households' ability to purchase goods and services from foreign suppliers and to supply services (e.g., accommodation) via DIPs; these developments complicate measurement since business surveys may not capture transactions between domestic households via foreign DIPs. - Governments and NPISHs - Governments and non-profit institutions serving households (NPISHs) use digital ordering and digital delivery as buyers and sellers and should be covered in exhaustive measures of digital trade. ### Non-monetary digital flows - Non-monetary digital flows refer to data and information flows exchanged without a monetary transaction (for example, social networking sites or search engines offering services in exchange for user data used to generate advertising revenues). - International banking depends on cross-border data flows to support services being provided; while international transactions in advertising or banking services can be captured in trade statistics, the supporting data flows are not captured. - Ongoing investigations aim to better understand and quantify non-monetary data flows given their importance. - Research in the context of the SNA revision concluded that services provided free of charge to end-users are already implicitly included in the value of goods and services in the current SNA production boundary. - For the time being, non-monetary digital flows are not in scope for digital trade. Paid transactions for data (e.g., sales of data sets) and trade transactions facilitated by data flows are included in measures of international trade and should be included in relevant components of digital trade where appropriate. ### Accounting principles - Accounting principles for recording digital trade generally follow BPM6 (IMF, 2009), IMTS 2010 (UN, 2011) and MSITS 2010 (UN et al, 2010). - Transactions that pass through DIPs require clarification, especially those facilitating services: - BPM6 (para 10.160) covers subcontracting/outsourcing and recommends that “the value of services exported and imported in the economy of the service arranger is recorded on a gross basis,” implying the arranger consumes the service and then supplies it to the customer. - Intermediation services provided by DIPs differ fundamentally from subcontracting: DIPs are often completely automated, do not take ownership of goods, nor render the services they intermediate. - The Handbook recommends recording only the intermediation fees, not the full value of the products being intermediated, in the accounts of DIPs—reflecting economic substance and consistent with recent BPM6 and SNA update research. ### Recommended reporting template - The Handbook provides a standardized reporting template (Table 2.1) to compile and present digital trade transactions, accommodating the multi-dimensional nature of digital trade. - The template includes the two main components: digitally ordered trade (item 2) and digitally delivered trade (item 3), and allows compilers to use the sources best suited to each component (e.g., ICT/e-commerce surveys for digitally ordered trade; services trade sources for digitally delivered trade). - The template includes an item for digitally ordered services trade (item 2.2) which can be obtained from ICT/e-commerce surveys that collect monetary values and additional breakdown questions. - The template allows for cases where compilers have information only on total digitally ordered trade or only on total digitally delivered trade, provided an estimate of double-counting (item 4) is available to subtract when aggregating to total digital trade. - The template can be expanded to include additional dimensions (e.g., link to (Services) Trade by Enterprise Characteristics (TEC/STEC), breakdowns by institutional sector, industry, firm size). Metadata on institutional sectors, industries, sizes of firms, etc. are important for comparability. - Addendum items: - Digital trade in services: total for digitally ordered and/or digitally delivered services. - Digitally deliverable services: compilers should generally be able to produce estimates without modifying existing sources by identifying within existing trade statistics the service categories that are digitally deliverable. - Reporting template structure (as provided): - Item | Total exports | Total imports - 1 Total digital trade | 2+3 minus 4 - 2 Digitally ordered trade | 2.1+2.2 - 2.1 Goods - 2.1.a of which: via DIPs - 2.2 Services - 2.2.a of which: via DIPs - 3 Digitally delivered trade - 3.a of which: via DIPs - 4 Digitally ordered and digitally delivered trade - 4.a of which: digital intermediation services - Addendum items: - A.1 Digital trade in services | 2.2+3 minus 4 - A.2 Digitally deliverable services | >3 ### Work on updating national accounts and balance of payments standards - The Handbook's conceptual framework aligns with SNA 2008 (UN, 2008a), BPM6 (IMF, 2009), IMTS 2010 (UN, 2011) and MSITS 2010 (UN et al., 2010); updates to these standards (e.g., any change in the production boundary) will be reflected in the measurement framework without impacting the statistical definition of digital trade. - At the time of preparing the Handbook, updates to national accounts and balance of payments standards (SNA 2025 and BPM7) were ongoing, with digitalization prominent in the research agenda; the updated SNA and BPM are to include common chapters addressing digitalization's impact on macroeconomic statistics. - The update process toward SNA 2025 and BPM7 provides clarifications useful to compilers even before new standards are in place. Relevant guidance notes include: - Digital intermediation services (GN C.4) - Clarifies difference between subcontracting and intermediated services where an intermediary arranges supply without rendering the service itself (includes DIPs). - Assimilates such intermediation services with services provided by agents and recommends recording DIP fees separately from the main transaction, under trade-related services. - Cloud computing (GN DZ.8) - Defines cloud computing services as “computing, data storage, software, and related IT services accessed remotely over a network, supplied on demand and with measured resource usage that allows charging on a pay-per-use basis”. - Recommends treating payments for software subscriptions as purchases of services; long-term licences for software should be considered fixed assets regardless of hosting. - International transactions in cloud computing should be recorded under computer services, as digitally ordered and digitally delivered trade as appropriate. - Non‑fungible tokens (NFTs) (GN DZ.10) - Defines NFTs as digital records hosted on a blockchain associated with a digital or physical asset that record rights assigned to their owner and are distinct from the associated asset. - Recommends recording NFTs based on rights conferred: - (a) NFTs conferring personal use and display rights should generally be recorded as consumption, although some may gain features of valuables. - (b) NFTs conferring commercial or other rights beyond personal use should be seen as contracts, licenses or leases. - (c) NFTs conferring full ownership of an associated asset should not be recorded as assets (presuming the asset itself is already recorded). - International transactions in NFTs for personal use (point (a)) are generally to be recorded in services, as digitally ordered and digitally delivered trade as appropriate. - Fintech (GN F.7) - Discusses implications of new financial products, services, technologies and access modes introduced by fintech. - Fintech activity and transactions are to be allocated within existing institutional sector, activity and product breakdowns and separately identified (with “of which” categories) where relevant; this applies to international trade in services. - Note: Transactions should be broken down by relevant product groupings (EBOPS 2010 for services and, for example, the Harmonized Commodity Description and Coding System (HS) or the Central Product Classification (CPC) for goods). Annex B of the Handbook provides examples to guide compilers in using the reporting template to record digital trade transactions. *Source: IMF, OECD, UNCTAD and WTO.* ### CHAPTER 2 — THE CONCEPTUAL FRAMEWORK FOR MEASURING DIGITAL TRADE ### CHAPTER 2 — THE CONCEPTUAL FRAMEWORK FOR MEASURING DIGITAL TRADE ### Definitions and scope - Crypto assets (GN F.18/GN DZ.2) - In 2023, BOPCOM and the AEG agreed on the treatment of non-liability crypto assets as non-produced non-financial assets and therefore excluding them from the scope of digital trade. - Recording of data (GN DZ.6) - Data that are produced and used in production for more than one year meets the SNA characteristics of an asset and, as such, should be capitalized in national accounts. - Data can be sold in a market transaction and international transactions in data are to be recorded in services, digitally ordered and digitally delivered as appropriate. - Definitions used for this Handbook - The terms “goods” and “merchandise” are used interchangeably to describe goods “which add to or subtract from the stock of material resources of a country by entering (imports) or leaving (exports) its economic territory” (UN, 2011). - Digital intermediation platforms (DIPs) in this Handbook exclude financial intermediation. - Digital intermediation services exclude financial intermediation. - The Handbook’s scope of goods and services reflects SNA 2008 (UN, 2008a), BPM6 (IMF, 2009), IMTS 2010 (UN, 2011) and MSITS 2010 (UN et al., 2010). ### Identifying digitally ordered transactions - Core definition (Handbook): - “The international sale or purchase of a good or service, conducted over computer networks by methods specifically designed for the purpose of receiving or placing orders”. - Key inclusion criteria (both must apply): 1. “over computer networks”, and 2. “by methods specifically designed for the purpose of receiving or placing orders”. - Scope clarifications: - Aligned with the 2009 OECD definition of e-commerce, but focused on international transactions. - Payment and ultimate delivery do not have to be conducted online. - Orders via internet, extranet or EDI messages are included. - Orders made by phone, fax or manually typed email are excluded from the OECD definition (see Box 3.1 for UNCTAD guidance on inclusion of email in some countries’ statistics). - Digital ordering covers websites/apps using online shopping carts or order forms whether on seller-owned sites or third-party marketplaces. - Orders via chat bots, virtual agents, and voice commands to virtual assistants meet the criteria and should be included. - Recurring transactions (subscriptions) initiated by a digital order: - All transactions associated with international digital orders placed in the current reporting period should be included in digitally ordered trade. - When firms lack information on the original ordering method for recurring payments, it may be necessary to estimate the share of current subscription income arising from digital orders (for example, based on the share of digital ordering among subscriptions initiated in the current period). - Framework contracts negotiated offline that give rise to subsequent online orders: online orders ensuing under the agreement should be included in digitally ordered trade. ### Measuring digitally ordered trade: sources, coverage and challenges - Nature of digitally ordered trade: - Consists of international e-commerce transactions involving buyers and sellers from businesses, households, government bodies, or NPISHs. - Most common flows: business-to-business (B2B) and business-to-consumer (B2C). - Measures of businesses’ e-commerce revenues and households’ e-commerce expenditures in a given economy will partially overlap domestically but are mutually exclusive and additive in the international trade context (either buyer or seller is outside the compiling economy). - Role of digital intermediation platforms (DIPs): - Presence of a DIP can obscure whether buyer and seller are resident in the same country for survey respondents. - A transaction between domestic parties may be reported as international e-commerce if the DIP is foreign-resident; conversely, a cross-border purchase may not be reported as trade if the DIP is resident domestically. - The DIP provides digital intermediation services to buyer and seller; these services should be recorded as digitally ordered and digitally delivered trade when the DIP is resident in a different economy from the buyer/seller. - Data source landscape (high-level) - No single source provides a holistic measure of digitally ordered exports and imports at the whole-economy level. - Compilers should identify and combine multiple sources to improve coverage (Figure 3.1 referenced for mapping sources and Table 3.2 for strengths/limitations). - Surveys are valuable because they can be designed to cover relevant institutional units, trade flows and reporting items and ensure conceptual alignment. - Practical measurement challenges: - Survey respondents, especially households, may struggle to identify and report international transactions when purchases go through DIPs. - Firms may lack information to link recurring payments to the original ordering channel. - Inclusion/exclusion of manually typed emails varies by country; when included, it is recommended to measure or at least ask whether reported amounts include orders placed via messages (UNCTAD guidance; see Box 3.1). ### Special cases and operational guidance - Financial, insurance and pension services (Box 3.2) - These services are within scope for digital trade when the customer applied online for the service (e.g., opening an account, subscribing to a contract). - Distinctions to consider: - Ongoing subscription fees initiated online are included. - Subsequent actions (e.g., instructing a trade that incurs an “action fee”) should be assessed separately according to whether the order was placed digitally. - Contract modifications requested offline (e.g., by phone) that establish a new contract/transaction are excluded from digitally ordered trade. - Recommendation: develop specific guidelines to help providers operationalize reporting for complex financial, insurance and pension services and share guidelines internationally. - Mobile roaming services (Box 3.2) - An individual using their home SIM to connect to a host country network purchases the service from their home telecommunications provider. - The transaction is between host and home country telecom providers and should reflect the amount charged by the former to the latter for the roaming service; this amount should be included in digitally ordered trade. - Transactions between affiliated enterprises (Box 3.2) - Private networks or proprietary computer systems used by affiliates to place orders constitute digitally ordered trade under the same principles that apply to unaffiliated enterprises. - UNCTAD guidance on e-commerce ordering channels (Box 3.1) - OECD excludes phone, fax and manually typed email as “specifically designed” ordering methods. - UNCTAD’s guidance and the UNCTAD Manual (2020) recommend collecting data on orders received or placed over the internet, including by email, to accommodate differing levels of technological development. - If manually typed messages are included, the value of such transactions should ideally be measured separately; if not possible, ask respondents whether reported amounts include orders placed via messages to indicate prevalence. ### Recommendations and practical considerations for compilers - Use multiple data sources and surveys to cover exports and imports involving different institutional sectors; combine sources where no single source is comprehensive. - Ensure survey design records and communicates coverage clearly (ordering channels included/excluded) to allow interpretation and comparability of statistics. - For recurring subscription payments where original ordering channel is unknown, consider estimating the current-period share arising from digital orders based on the share of digital ordering among subscriptions initiated in the current period. - Provide specific, shared guidance for complex service providers (notably financial, insurance and pension services) to improve operational reporting and international comparability. - For orders mediated by DIPs, develop methods to: - Identify residency of buyer and seller despite third-party/platform involvement. - Record digital intermediation services provided by the DIP when the DIP is resident in a different economy from buyer/seller. *Handbook on Measuring Digital Trade — CHAPTER 2 — THE CONCEPTUAL FRAMEWORK FOR MEASURING DIGITAL TRADE* ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### Institutional coverage and data sources - Figure 3.1 maps institutional sectors (Businesses; Households; Government / Non-profit institutions serving households (NPISH); All institutional sectors) against source types and conceptual alignment. - Source types listed with their coverage notes: - Information and communications technology (ICT) surveys — Full coverage / conceptual alignment (depending on survey design). - Core business surveys — Full coverage / conceptual alignment (depending on survey design). - Multinational enterprise (MNE) surveys — Partial coverage – MNEs only; conceptual misalignment – may include sales through local subsidiaries/affiliates outside the compiling country. MNE surveys tend to focus on sales (exports) and not to cover businesses’ purchases (imports). - Value added tax (VAT) returns — Partial coverage – only VAT registered businesses; may exclude businesses which sell online through channels other than their own website/webshop (e.g., via online marketplaces or EDI); possible over-coverage – may include offline sales by businesses selling online. - Card payments — Partial coverage – only payments made by card; possible over-coverage – may include payments made on corporate/business cards or payments made on personal cards for business purposes. Breakdown into goods, services, digitally delivered services may be possible based on merchant category codes. - Customs declarations — Partial coverage – only goods above relevant customs/statistical thresholds. - Low-value trade estimates — Partial coverage – only goods below relevant customs/statistical thresholds. - Exports (X) / Imports (M) can be captured across sources; digitally ordered trade broken down by: - Goods - Of which: via DIPs - Services - Of which: via DIPs - Digitally ordered and digitally delivered trade - Note: Section references indicate where further details on each source can be found in this chapter. ### Businesses as primary actors in digitally ordered trade - Empirical evidence and estimates: - UNCTAD estimates that around 80 per cent of businesses’ e-commerce sales (by value) are made to buyers that are also businesses (UNCTAD, 2021c). - Box 3.3 evidence highlights: - On average across OECD countries, nearly 30 per cent of businesses received orders over computer networks in 2022. - In the same year, around 20 per cent of individuals sold goods or services online. - In Canada, 12 per cent of persons aged 15 years or older reported earning money online in 2020. The average earning from online activities was CAD 2,700 (around US$ 2,000). By comparison, in the following year the average e‑commerce sales value was CAD 3.7 million across all businesses and over CAD 500,000 for small enterprises (Statistics Canada, 2022a). - In Japan in 2021, business‑to‑business (B2B) e‑commerce transactions amounted to over YEN 370 trillion, business‑to‑consumer (B2C) e‑commerce to almost YEN 21 trillion, and consumer‑to‑consumer (C2C) e‑commerce to just YEN 2.2 trillion (METI, 2022). - In 2013, C2C e‑commerce accounted for only 1 per cent of the total value of e‑commerce sales in the Republic of Korea (Statistics Korea, 2014). - Conclusion from evidence: businesses are the main actors in e‑commerce and therefore, by extension, in digitally ordered trade. - Implications: - Digitally ordered trade involving businesses as both sellers (exports) and buyers (imports) should generally be the highest measurement priority. - Compilers should assess the prevalence and importance of cross-border e-commerce transactions involving government units, NPISHs, and households to set priorities and ensure exhaustiveness and representativeness. ### Digitally ordered trade in goods and customs-based measurement - Rationale: - Goods account for over three quarters of global trade (WTO, 2022). - Limited evidence suggests that around two-thirds of e-commerce sales relate to goods (see Section 3.3). - Main sources and complementarities: - Customs declarations are the main source for merchandise trade statistics. - Customs data, when complemented by low-value trade estimates, provide a holistic measure of exports of goods from, and imports of goods to, the compiling economy. - Implementing identification of digitally ordered shipments in customs systems offers the possibility of measuring a large component of digitally ordered trade across institutional sectors and integrating with international merchandise trade statistics (see Section 3.2.2). - Limitation: - Customs-based measures of digitally ordered trade in goods need to be complemented with figures on digitally ordered services imports derived from other sources. ### Combining sources and microdata linking - Key implication: combination of data sources is necessary to produce whole-economy statistics on digitally ordered trade. - Example approaches: - Card payments data might offer a robust measure for household expenditure on digitally ordered imports complementary to business survey measures (see Section 3.2.1). - Microdata linking: integrating goods and services trade data with responses from business ICT surveys to: - Identify exporting businesses that make at least some sales via e-commerce. - Identify importing businesses that use e-commerce to purchase some inputs. - Use total imports and exports known from trade sources combined with ICT survey information to estimate the portion of those trade flows that is digitally ordered. ### Survey sources: overview and use for measuring digitally ordered trade - Conceptual alignment: - Definition of digitally ordered trade transactions aligns with e-commerce transactions except that digitally ordered trade is confined to transactions between residents and non-residents; thus surveys measuring e-commerce can form the basis for measuring digitally ordered trade. - Focus to date: - Most measurement efforts have focused on businesses and households. - Business surveys can ask about e-commerce sales revenues and expenditures; household surveys focus on e-commerce spending and sometimes household online income from selling items or services. ### Business surveys — ICT surveys and enhancements - Coverage and prevalence: - Nearly 80 countries worldwide have undertaken ICT/business e-commerce collections (UNCTAD, 2023). - Annual business ICT surveys are legally mandated for EU member states and also take place in other European Statistical System participants. - Annual or biennial surveys occur in most other OECD countries and in Brazil (which submits statistics to the OECD database on ICT access and usage by businesses). - Recurring collections exist in many Asian economies including China, Indonesia, Malaysia, the Philippines, Singapore and Thailand; statistics available in the UNCTAD database of core indicators on ICT usage in business. - Core indicators and adoption: - Core ICT indicators include the share of businesses making e-commerce sales and the share purchasing inputs through e-commerce. - Core indicators have been officially adopted by countries through endorsement at the UN Statistical Commission. - Monetization of e-commerce in surveys: - Logical enhancement: ask businesses that engage in e-commerce for monetary values of income from e-commerce sales and expenditure on e-commerce purchases, or as percentage shares of total sales/expenditure. - Many countries collect some value information; EU member states have collected data on the value of turnover from e-commerce orders since 2012. - As of 2023, only a relatively small number have published monetary figures on the value of e-commerce sales (UNCTAD, 2023). - Typical supplementary questions: - Customers to whom the business sells (businesses, government, consumers). - Sales channels used (own websites/apps, third-party websites/apps/marketplaces, EDI messages). - Whether the business has made e-commerce sales to customers abroad. - Breakdown of total e-commerce sales value into domestic and international components (percentage shares or monetary amounts) — enabling derivation of digitally ordered exports. ### Survey design guidance and examples - Guidance for DIPs and double-counting: - Example practice: digital intermediation platforms (DIPs) that receive internet orders on behalf of other entities are instructed to enter only the fees earned on the transaction to mitigate double counting (e.g., hotel sales reported by hotels and by booking platforms). - Inclusion of specific guidance in questionnaires for respondents in lines of business where the digital ordering concept is not straightforward is recommended (e.g., DIPs; businesses providing financial, insurance and pension services; affiliated enterprises). - Geographic breakdowns: - Several countries request geographic detail on e-commerce sales abroad: - EU surveys ask businesses to delineate e-commerce sales to the respondent’s own country, other EU member states and the rest of the world. The cross-border breakdown was included as a mandatory breakdown for the first time in 2021. - Canada’s Survey of Digital Technology and Internet Use collects shares of e-commerce sales revenue from different geographic regions and, in some cases, specific countries; published 2021 results show 20 per cent of e-commerce sales by businesses in Canada were to customers abroad, with over three-quarters of these orders (by value) going to customers located in the United States. - The United Kingdom ONS 2021 Digital Economy Survey asks respondents to break down e-commerce turnover by geographic regions and nests this within a top-down question on turnover from sales to customers outside the United Kingdom. - Breakdown by product type and channels: - Important breakdowns include sales of goods, non-digitally delivered services and digitally delivered services — crucial to estimate overlap between digitally ordered trade and digitally delivered trade and to avoid double counting (reporting template item 4; see Table 2.1 in Chapter 2). - ONS collected a separate breakdown isolating turnover via “online marketplaces” (DIPs), relevant to reporting template items 2.1.a and 2.2.a. - Country examples and outcomes: - Malaysia (Department of Statistics Malaysia): Box 3.4 example shows total value of e-commerce sales can be collected directly as monetary value or as a percentage of total sales revenue; respondents provide domestic vs abroad breakdown as shares adding up to 100 per cent; results show total value of business e-commerce sales increasing rapidly over time and share of sales to customers abroad increasing. - Spain (INE): published results show business e-commerce purchases in Spain amounted to EUR 222 billion in 2020; total business e-commerce sales in Spain were EUR 275 billion in the same year. Spending via e-commerce amounted to 23 per cent of purchases across all businesses, and 45 per cent of purchases by businesses which used e-commerce to buy goods and services. EUR 53 billion of e-commerce purchases by businesses in Spain in 2020 were from sellers/suppliers abroad, the majority in other EU member states. - Trade-offs: - Each additional survey question increases respondent burden and may lower response rates. - Statistical compilers must balance the need for detailed information with managing respondent burden and response rates. *Source: CHAPTER 3 — DIGITALLY ORDERED TRADE, Handbook on Measuring Digital Trade* ### Chapter 2) offers flexibility, allowing countries to ### hmdtea - Chapter 2) offers flexibility, allowing countries to ### Survey examples and questionnaire design - Malaysia: 2020 Survey on Usage of ICT and E‑commerce by Establishment — detailed question modules capturing: - Total number of persons engaged (full-time and part-time) (question codes: 310002, 310003, 310004, 310011, 310016). - Computer and internet use for business purposes (questions 4.1, 5.1, 5.3, 5.4). - Platforms used for sales/purchases via internet (question 6.2; examples: Social media, Designated private network (e.g., extranet, EDI), Payment gateway, Website owned by establishment, Mobile app). - Whether establishment received orders via e‑commerce (question 6.3; include: websites, specialised internet marketplaces, extranet, EDI over the internet, Internet-enabled mobile phones; Exclude: orders submitted via email). - Methods of payment used for internet transactions (question 6.3; example: Cash on delivery). - Income from e‑commerce sales and instructions for reporting (question 6.5; unit RM; note: for orders received on behalf of other organisations, report only fees or commissions). - Percentage breakdown of e‑commerce income by customer type (B2B, B2C, B2G) and market (Domestic, International) (questions 6.6, 6.7, 6.8). - Platform breakdown of e‑commerce income by where orders are placed (questions 6.8, 6.9). - Request for country names with highest international e‑commerce income (question 6.9). - European Union: 2021 European Community Survey on ICT Usage and E‑commerce in Enterprises (Eurostat, 2021b) — harmonized module: - WEB sales definition (customer places order on website or through an app). - Question B2: value of web sales in 2020 (national currency, excluding VAT) OR percentage of total turnover (approximation acceptable). - Question B8: percentage breakdown of web sales in 2020 by geographic area — Own country, Other EU countries, Rest of the world; Total 100 %. - Note: harmonized survey only breaks down “web sales” into domestic and international; around 65 per cent of e‑commerce turnover in EU27 comes from EDI‑type sales, and EDI‑type sales breakdown is less widely available. - Canada: 2021 Survey of Digital Technology and Internet Use (Statistics Canada): - Question 21: total gross sales conducted over the internet in 2021 (rounded to nearest CAN$). - Question 22: percentage of gross sales made over the internet in 2021. - Question 26: percentage of internet gross sales by region in 2021 — Canada; United States; Mexico; Other Latin America and the Caribbean; China; Other Asia; The European Union; The United Kingdom; Other regions; Total 100 %. - United Kingdom: 2021 Digital Economy Survey (Office for National Statistics) — collects: - Turnover from e‑commerce sales to customers outside the UK (value in £) and breakdowns by area, by platform (own website/app, own social media, online marketplace, EDI, other platforms), by product/service type (goods, digitally delivered services, non‑digitally delivered services). - Expenditure on e‑commerce purchases from suppliers outside the UK and analogous breakdowns. - Spain: Instituto Nacional de Estadística Survey on the use of Information and Communication Technologies and Electronic Commerce in companies 2020: - Total amount of purchases of foreign goods and services in 2020 (excluding VAT) reported in €. - Purchases by electronic commerce through web/mobile applications and by EDI; estimated percentage of total purchases via each mode and geographic breakdown (Spain; Other EU countries; Rest of the world; Total 100 %). ### Key findings and measurement issues - Coverage variation: - Business ICT surveys vary significantly by industry coverage and firm size (for example, EU surveys commonly exclude firms with fewer than 10 persons while many other countries include microenterprises). - Common industry omissions: “Agriculture, Forestry, and Fishing” (ISIC Rev.4 section A), “Mining and Quarrying” (B), and “Public Administration and Defence” (O). Under the EU model, additional out‑of‑scope sectors include “Finance and Insurance” (K); “Education” (P); “Human Health and Social Work” (Q); “Arts, Entertainment, and Recreation” (R); and most of “Other service activities” (S). - These exclusions affect exhaustiveness and comparability of business e‑commerce and digitally ordered trade figures across economies (UNCTAD, 2023). - Importance of EDI: - Around 65 per cent of e‑commerce turnover earned by businesses in the EU27 comes from EDI‑type sales, making EDI coverage critical for complete measurement. - Respondent selection: - Business ICT surveys are often addressed to businesses’ IT departments; surveys should clearly state that respondents may need input from colleagues in other departments (e.g., sales/accounting) when answering questions on sales via digital ordering. ### Policy and statistical recommendations - Ensure survey coverage and methods: - Statistical compilers are recommended to ensure that the coverage of business ICT surveys, and the methods and estimations applied to responses, are sufficient to derive digitally ordered trade estimates representative of all businesses. - Any exclusions and representativeness limitations must be communicated to users and documented in metadata. - Flexibility in data collection: - Countries may report key items, such as total digitally ordered trade and the sub‑component relating to digitally delivered services, without imposing the need to collect all breakdown items (flexible reporting approaches). - Allow approximations: - Where precise figures are not available, allow reporting of percentages or approximations (explicitly noted in Eurostat and other surveys). ### Key statistics and examples preserved from source - Malaysia survey question and code references: 310002, 310003, 310004, 310011, 310016, 310036, 310043, 310050, 310056–310061. - Payment method example: Cash on delivery. - Currency units shown in examples: RM, CAN$, £, €. - EU‑specific statistic: "around 65 per cent of e‑commerce turnover earned by businesses in the EU27 comes from EDI‑type sales." - Figure indicators in source excerpts: - Figure 3.2: shown values “12%”, “12%”, “14%”; axis labeled "RM billions, current prices". - Figure 3.3: category "Abroad 20%" and axis labeled "CA$ billions". - Figure 3.4: axis labeled "EUR billions, current prices"; years shown 2016–2020. *Source: Department of Statistics Malaysia, Survey on Usage of ICT and E‑commerce by Establishment 2020; excerpts from Handbook on Measuring Digital Trade (Chapter 3).* ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### Core business surveys - Some countries measure the total value of e-commerce sales through questions included on “core” macroeconomic business surveys (examples: Annual Survey of Philippine Business and Industry; Annual Services Industry Survey in Singapore; separate annual surveys of manufacturing, services, retail and wholesale businesses in the United States). - These surveys: - Measure “sales from e‑commerce transactions” or “e‑commerce revenue of the services sector”, sometimes broken down by ISIC Rev.4 industry sections and by customer type (business or consumer). - Do not typically measure international e‑commerce transactions directly, but total e‑commerce values can be combined with import/export information from the same businesses to derive first estimates of digitally ordered trade. - Could be expanded in a top‑down approach similar to business ICT surveys to measure digitally ordered trade and integrate measures with gross value added and GDP. - Advantages of using routine business activity surveys: - Wider coverage of industries and firm sizes. - Closer integration with major economic aggregates (gross value added of the business sector and GDP). - Combined approach: - Mainstream enterprise surveys gathering headline digitally ordered sales/purchases can be used alongside detailed ICT surveys to combine population coverage with detailed breakdowns. - Trade in services surveys: - Possible to add questions on digital ordering to international trade in services surveys (would not cover digitally ordered goods). - Could provide insights on overlap between digital ordering and digital delivery. - Recommendation: - Statistical compilers should explore mainstreaming additional relevant questions in core business surveys used to derive structural business statistics and/or in international trade in services surveys. - Sample ICT survey questions and experiences can serve as a starting point for developing questions for other business surveys. ### Multinational enterprise (MNE) surveys - MNE surveys can collect data on digital ordering, digital delivery, and digital intermediation platforms. - Scope and contribution: - MNEs are a subset of businesses but can account for a large share of trade: in the United States, over 90 per cent of services trade and a majority of goods trade was driven by MNEs (Bruner and Grimm, 2019). - Surveys of MNEs therefore have the potential to measure a considerable portion of digital trade. - Implementation: - MNEs can be included in general balance‑of‑payments surveys (e.g., international trade in services surveys) or surveyed separately. - Large case units (LCUs) in national statistics offices or central banks often manage data collection on MNEs and may collect balance sheets, income statements, and sales to domestic and foreign customers that contribute toward measures of digital trade. - Question content and benefits: - MNE surveys may ask if firms engage in online sales and request shares of sales revenues that are digitally ordered and/or digitally delivered. - Resulting data can be used alone for partial measures or combined with trade data (aggregate or microdata) for more exhaustive digital trade statistics. - Surveys can collect information on trade in goods and services in the same instrument. - Box: United States BEA experience (2019 Benchmark Survey of U.S. Direct Investment Abroad and 2022 Benchmark Survey of Foreign Direct Investment in the United States) - Questions collected shares of: 1) services sales that are digitally delivered; 2) services sales that are digitally ordered; and 3) goods sales that are digitally ordered. - Response format: - Checkboxes indicating percentage ranges: 0%, 1-24%, 25-49%, 50-74%, 75-89%, 90-99%, 100%. - Respondents indicated the information basis: Accounting records or Recall/general knowledge of operations. - Respondents were asked to provide reasonable estimates based on informed judgement, sampling techniques or prorations and to indicate the basis for responses. - Findings among responding US parent companies: - Overall share of services that were digitally ordered was 25 per cent (estimated by multiplying the midpoint of the percentage range by the reported sales for each respondent, then adding up all respondents). - Share of goods that were digitally ordered was 19 per cent. - Shares for foreign affiliates were between 30 and 35 per cent. - Responses varied across industries (digital ordering most prevalent for services in Information Services and Professional, Scientific, and Technical Services; for goods in Manufacturing and Retail Trade). - Three in four US parent companies and two in three affiliates reported responding based on recall or general knowledge rather than accounting. - Just over half of US parents did not respond to these questions. - Operational notes: - BEA followed up directly with many respondents to ensure accurate responses and is exploring ways to account for non‑responses and for MNEs not asked the digital economy questions. - Definitions provided to respondents: - Digitally ordered: sale of a good or service conducted over computer networks by methods specifically designed for receiving or placing orders, negotiating terms of sales or price (covers electronic data interchange, the internet, mobile devices, or any other online system). - Digitally delivered services: services delivered remotely over information and communications technology networks — over voice or data networks, including the internet, or in an electronically downloadable format. ### Household surveys - Role and focus: - Household ICT access and usage surveys measure proportions of individuals purchasing and selling goods and services online; they tend to focus more on e‑commerce expenditure than on value for cross‑border transactions. - Eurostat 2021 European Community Survey on the Use of ICT in Households and by Individuals: - Respondents asked to estimate total purchases via websites or apps over the three months prior to the survey, either as an exact amount in national currency or by choosing spending bands: - Less than 50 euro - 50 to less than 100 euro - 100 to less than 300 euro - 300 to less than 500 euro - 500 to less than 700 euro - 700 to less than 1000 euro - 1000 euro and more - Don’t know - Respondents asked from whom they bought via website or app in the last 3 months (tick all that apply): 1. National sellers 2. Sellers from other EU countries 3. Sellers from the rest of the world 4. Country of origin of sellers is not known - Eurostat results: - Most common spending band: EUR 100 and 299 over the three months prior to survey (varies by country). - 21 per cent of individuals (aged 16-74) had made online purchases from sellers outside their country of residence. - Limitation: - Questions on seller location and value of online spending are separate, so the survey does not provide the share of online spending attributable to purchases from abroad. - Canadian Internet Use Survey 2020: - Collected specific amounts spent on various online purchases of goods and services, allowing categorization into goods, digitally delivered services, and other services to avoid double counting when calculating total digital imports. - Did not distinguish between domestic and international e‑commerce in 2020. - 2018 edition asked if online orders of physical goods were “delivered from” merchants in Canada, the United States, other countries, or merchants of unknown country of origin; the monetary amounts were not broken down by seller location. - Respondent challenges: - Difficulty delineating domestic vs international transactions (example: Amazon.ca orders reported as local by some respondents and as international by others). - Household earnings: - 2020 survey included questions on “online work” to measure money made online by individuals, but the international dimension was not collected. - Recommendations and observations: - Household surveys can yield meaningful results on the share of digital ordering in household expenditure and on online earnings. - Ideally, household surveys should collect whether the purchased/sold product is a good, a digitally delivered service, or a service delivered via another channel. - Breaking down household e‑commerce spending by product type can show where spending implies imports of digitally ordered and digitally delivered services (e.g., music streaming, online gaming, online storage). - Product details can inform whether household online earnings imply digitally ordered exports (e.g., platform‑based peer‑to‑peer accommodation in economies with limited domestic tourism). - Other household surveys could be expanded to collect values of international digitally ordered transactions similarly to business surveys, but respondents often struggle to identify cross‑border transactions accurately and consistently. - Survey questions and methods are still at an early stage; further design, testing, and experimentation are needed. ### Key statistics and data points preserved from the chapter - Over 90 per cent of services trade in the United States was driven by MNEs (Bruner and Grimm, 2019). - BEA survey findings: - Overall share of services digitally ordered among responding US parent companies: 25 per cent. - Share of goods digitally ordered among responding US parent companies: 19 per cent. - Shares for foreign affiliates: between 30 and 35 per cent. - Response basis: Three in four US parent companies and two in three affiliates used recall/general knowledge rather than accounting. - Just over half of US parents did not respond to the digital economy questions. - Eurostat household result: 21 per cent of individuals (aged 16-74) had made online purchases from sellers outside their country of residence. - Eurostat spending bands for last 3 months (euros): Less than 50; 50 to less than 100; 100 to less than 300; 300 to less than 500; 500 to less than 700; 700 to less than 1000; 1000 and more. - Canadian Internet Use Survey 2020 used ranges for physical goods spent in past 12 months (CAD): - Less than $200 - $200 to less than $500 - $500 to less than $1,000 - $1,000 to less than $5,000 - $5,000 or more - Canadian survey collected itemized dollar amounts for digital goods and services (music downloads/streaming subscriptions, video downloads/streaming subscriptions, E‑books, podcasts, online newspapers/magazines, digital gift cards for online redemption, online gambling, online gaming, online data‑storage services, online courses, other applications/software/subscriptions, other digital goods or services) and peer‑to‑peer accommodation total. *Source: CHAPTER 3 — DIGITALLY ORDERED TRADE, Handbook on Measuring Digital Trade.* ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### Household surveys and measuring digitally ordered expenditures and earnings - Household expenditure and international travel surveys can be adapted to identify the share of expenditures on accommodation and (separately) travel services purchased abroad that were digitally ordered; this can help identify and quantify potential underestimates (see Box 3.13). - Conventional household income surveys could ask whether households provided short-term accommodation services via digital intermediation platforms (DIPs) and the income generated; such questions would provide an order of magnitude and an upper estimate even if they do not initially differentiate between services provided to residents and to non-residents. - Most, if not all, e-commerce sales by individuals/households would not take place without DIPs; transactions involving DIPs are discussed further in Chapter 5. ### Box 3.12 — Measuring online earnings by individuals in Canada (survey questions and results) - 2020 Canadian Internet Use Survey question asked respondents: "During the past 12 months, how much did you personally earn by doing the following activities online?" with itemized lines including: - Selling physical goods online that you built or created _____$ - Selling services via online bulletin boards _____$ - Providing platform‑based peer‑to‑peer accommodation services _____$ - Providing platform‑based peer‑to‑peer ride and delivery services _____$ - Providing other platform‑based peer‑to‑peer services _____$ - Online freelancing _____$ - Crowd‑based microwork _____$ - Earning income through online advertisements and sponsored content _____$ - Other activities _____$ - 2022 survey wording collected an overall amount of online earnings from these activities with explanatory text on gig economy examples (e.g., Uber, Lyft, TaskRabbit, Upwork, Guru, Fiverr, Freelancer). - Empirical results from 2020: - Average earnings from online activities: CAD 2,700 (around USD 2,000). - "Selling physical goods online that you built or created": 7 per cent of those aged 15 or over in Canada, with average earnings of almost CAD 1,700. - "Selling services via online bulletin boards [such as eBay or Kijiji]": 4 per cent, CAD 1,500. - "Online freelancing": 0.7 per cent, around CAD 10,000 on average. - "Other activities": 2.3 per cent and around CAD 19,000 on average. - Classification and coverage notes: - It is unclear which activities count as earnings from e‑commerce sales relevant for digitally ordered trade. - Income from selling accommodation, ride and delivery, or other services through online platforms would clearly meet the definition of digital ordering. - Many online sales of physical goods by individuals are likely via digital intermediation platforms such as eBay or Etsy; manually typed messages and emails used for sales are not included as e‑commerce in Canada. ### Surveys of government units and NPISHs - Institutional units beyond businesses and households (government units and non-profit institutions serving households — NPISHs) can engage in e-commerce and digitally ordered trade as buyers or sellers and should be covered for exhaustive measures. - Few ICT-usage surveys target these institutional sectors; those that exist focus on digitalization of processes (e.g., e-procurement) rather than transaction values. - Alternative data sources (e.g., government budgetary management reporting systems) may be necessary to collect relevant information for these sectors. - In most economies, business and household transactions likely constitute the majority of digitally ordered trade; coverage of government and NPISHs may have lower priority but should be considered where they could represent statistically meaningful volumes. ### Customs declarations and other administrative/private sources for digitally ordered trade - Customs records underpin international merchandise trade statistics; identifying digitally ordered shipments within customs sources permits measurement of digitally ordered imports and exports integrated with merchandise trade statistics. - The World Customs Organization (WCO), with public and private sector actors, is working to better identify and monitor digitally ordered trade in customs records via improved electronic identification (e.g., S10 bar code for postal items) and special declaration forms for e-commerce orders. - The WCO’s "Framework of Standards" on cross-border e-commerce provides structural guidance on measuring e‑commerce (digitally ordered) transactions and aims to establish global standards across the e‑commerce supply chain (risk assessment, clearance/release, revenue collection, border cooperation). - Examples of national efforts: - China Customs: using specific customs procedure codes and a Cross‑border E‑commerce Clearance System (CBECS) with simple declarations that combine and cross‑validate orders, logistics and payment data; e‑commerce platforms declare summarized data afterward to customs; complementary methods use postal agency sampling and big data methodologies to estimate B2C and B2B cross‑border e‑commerce (see Box 3.15). - Türkiye: TurkStat and the Ministry of Trade developed a methodology using traditional customs records with an added specific field to identify digitally ordered transactions, electronic customs declarations and postal data (see Chapter 6). ### De minimis and low-value international trade estimates — relevance to digitally ordered trade - Digital ordering has driven rapid growth in international parcel shipments (example: Japan consignments rose from 12.3 million in 2010 to 63.4 million in 2020). - Many shipments are low value and fall below customs de minimis thresholds (minimum value, weight, size and/or amount under which duties or taxes are not collected and details are not required). - Digitally ordered trade below de minimis thresholds will not be covered by customs-declaration-based measures and therefore must be separately measured or estimated for representative statistics. - De minimis thresholds vary widely across countries: - Example OECD range: GBP 15 (around US$ 17) in the United Kingdom to US$ 2,000 in the United States. - In most cases, de minimis trade amounts to around 1-3 per cent of total trade, but can reach over 15 per cent for some economies. - Around half of OECD countries, and several non‑OECD countries, produce measures or estimates of de minimis or low value trade for balance-of-payments and merchandise trade statistics; sources include national postal services, customs administrative reports, card payment information, or estimation models. - Cautions and methodological notes: - There is a strong correlation between growth in de minimis transactions and growth in digital ordering, but not all de minimis trade is digitally ordered; careful use of de minimis estimates is required. - Postal-delivery-provider information can provide robust low-value/de minimis estimates if it covers major postal and courier providers and all transport modes. - Country examples of low-value regimes: - Republic of Korea: - Imports for personal use with value less than US$ 150 are exempt from tariff and VAT (de minimis). - Goods with value between US$ 150 and US$ 2,000: "simplified import declaration" with 57 fields (low-value trade). - Goods over US$ 2,000: "General Import Declaration" with 69 fields and full documentation requirements. - European Union: - From 1 January 2021, goods up to EUR 150 may be declared using a customs declaration requiring one third of the data compared to a standard declaration. ### Box 3.13 — Compiling digitally ordered travel transactions in Italy (survey results) - Bank of Italy face-to-face border survey (since 1996) gathers inbound and outbound international tourism metrics; since 2016 specific questions ask about online purchases of "all‑inclusive" travel packages, online booking of accommodation, and channels used to book accommodation online. - Trends (period 2016‑21): - Expenditure on "all‑inclusive" trips purchased or booked online: - Residents in Italy: increased from 14 per cent to 25 per cent. - Non‑residents: increased from 18 per cent to 30 per cent. - Online booking of accommodation: - Residents: increased from 43 per cent to 49 per cent. - Non‑residents: increased from 66 per cent to 73 per cent. - Figure 3.7 (as described): shows shares of travel packages purchased online and shares of accommodation booked online for residents and non‑residents during 2016–2021. ### Box 3.14 — WCO Framework of Standards on cross‑border e‑commerce (measurement aims) - Framework grounded in eight guiding principles (Luxor Resolution 17, 2017); Principle V – Measurement and Analysis underpins Standard 15: Mechanism of Measurement, requiring customs administrations to capture, measure, analyse and publish e‑commerce statistics consistent with international statistical standards. - Implementation aims include: - Establishing common terminologies and reliable mechanisms to accurately measure and analyse cross‑border e‑commerce in cooperation with international organizations and stakeholders. - Using data analytics and existing capabilities of international organizations, e‑vendors/e‑platforms, and other stakeholders to generate trends and analysis for evidence‑based decision‑making. - Establishing mechanisms and supporting legal frameworks to capture data at item level while implementing simplified clearance processes (e.g., consolidated simplified summary declaration). ### Box 3.15 — Measuring cross‑border merchandise e‑commerce using customs data in China (methods) - China Customs innovations: - Specialized Cross‑border E‑commerce Clearance System (CBECS) with specific customs regime codes for B2C cross‑border e‑commerce; simple declarations combine and cross‑validate orders, logistics and payment data; e‑commerce platforms submit summarized data afterward. - Use of e‑commerce platform order information, cross‑border logistics records, cross‑border payments and big data methodologies to develop statistical estimates for cross‑border e‑commerce. - For B2C mail parcels and courier deliveries not cleared via CBECS, pilot surveys and sampling methods determine the proportion of e‑commerce postal parcels. - For B2B transactions, exporters are encouraged to declare whether goods were ordered via e‑commerce for future statistical survey validation. ### Box 3.16 — Low‑value trade estimations in the United States (methodology) - United States de minimis and low‑value thresholds and estimation approaches: - Thresholds: US$ 2,500 for exports; between US$ 800 and US$ 2,000 for imports. De minimis trade is distinct from de minimis threshold (certain imports below US$ 800 exempt from duties and some customs procedures). - Exports: estimated in two parts: - Low‑value trade delivered by small package couriers: use a "courier factor" equal to ratio of total low‑value trade to total high‑value trade by small package couriers; multiply this factor by value of high‑value trade delivered by small package couriers. - Non‑courier low‑value trade: use a country‑specific factor multiplied by each country’s trade from prior (or current, if available) month (Canada handled separately under United States‑Canada Data Exchange). - Imports: mostly based on low‑value import data (excess electronically filed data omitted from original statistics due to aggregation), supplemented with estimates of low‑value data filed via paper, estimates of low‑value trade transported by courier, and data on low‑value trade transacted within foreign trade zones. *Source: IMF — CHAPTER 3 — DIGITALLY ORDERED TRADE (Handbook on Measuring Digital Trade).* ### Box 3.16 provides a further example of differentiation ### Box 3.16 provides a further example of differentiation ### Differentiation between de minimis and low-value trade - Low-value trade regimes normally collect the information needed to accurately track and measure low-value trade and, in many cases, could be enhanced to gather information on digitally ordered low-value transactions as well. - Such regimes help distinguish de minimis thresholds from low-value e-commerce flows for statistical and administrative purposes. ### Administrative and customs data examples - Türkiye (since 2012): uses electronic trade customs declarations to measure low value digitally ordered trade. Declarations are issued electronically by authorized express airline cargo companies and help to expedite customs processes. - Upper limits for eligibility for electronic declaration: EUR 15,000/300 kilogrammes for exports and EUR 150/30 kilogrammes for imports. - Republic of Korea: registered e-commerce companies are subject to special customs reporting which reduces paperwork and expedites clearance (and possibly offers some duty and tax exemptions). - Australia: goods arriving destined for a private individual, valued at less than AUD 1000 and ordered through an online supplier, are subject to goods and sales tax (GST) which must be paid by the supplier. ### VAT returns data (links to Box 3.18) - Business VAT returns include the value of a business’s total sales and purchases but typically do not break down sales by channel (online vs offline). - If businesses selling online can be identified within VAT returns, their VAT-reported figures can be used to estimate e-commerce sales and digitally ordered trade. Box 3.18 — Measuring cross-border e-commerce from webshops in the Netherlands (summary) - Statistics Netherlands (CBS) used Dutch VAT returns filed by foreign EU companies to measure expenditure by Dutch consumers in non‑Dutch “webshops”. - Mandatory VAT returns thresholds in the EU: EUR 35,000 or EUR 100,000 per year, depending on the EU member state. - Identification process: - Combined VAT returns with Bureau Van Dijk ORBIS database to select enterprises engaged in retail. - Matched records using company names; significant editing required to avoid false negatives (punctuation, abbreviations). - Used big data analytical techniques with University of Amsterdam and Leiden University to improve linking. - Paired company overview with web scraping to identify webshop websites (shopping cart presence); largest foreign companies re-checked manually. - Manual checks indicated measurement errors of approximately 5 per cent of turnover; these errors can be reduced by training algorithms via machine learning. - Results for 2016: - Dutch consumers spent over EUR 1 billion (excluding VAT) on products sold by foreign EU webshops in 2016, an increase of 25 per cent relative to 2015, and a value six times higher than that previously recorded by demand‑side consumer surveys. - More than half of online purchases were from webshops in Germany, followed by the United Kingdom, Belgium and Italy. - Clothing and shoes were the most common items purchased. ### Australia’s vendor collection model for GST on low-value imports (Box 3.17 summary) - Implemented 1 July 2018: requires certain non-resident suppliers (including platforms, merchants and re-deliverers) to register for, apply and collect GST on low‑value imported goods sold to consumers in Australia. - Model exclusions: - Does not apply to business customers in Australia. - Does not apply to goods valued at over AUD 1,000 (the de minimis value in Australia). - Intent: target digitally ordered imports to level the playing field for domestic businesses previously disadvantaged by non‑resident online sellers not charging GST. - Mechanism: for goods arriving in Australia valued at less than AUD 1,000 and digitally ordered by households from a non-resident supplier, the supplier is expected to pay GST directly to the Australian Taxation Office (ATO). - Compliance and revenue: - High compliance among major platforms; suppliers registered and completed GST collection and remittance obligations. - ATO collections from low‑value imported goods: - AUD 225 million in the 2020‑21 Australian financial year. - AUD 161 million in the 2018‑19 financial year. ### Card payment data: potential and limitations - Data holders: banks (card issuers), global payment processing networks (Visa, Mastercard), specialist online payment processors (Alipay, PayPal), and digital intermediation platforms (DIPs). - Card payment data components useful for measuring digitally ordered trade: - Transaction amount. - Indicator for “card-absent” (or “card-not-present”) transactions when a card is used online. - Merchant outlet country (usually available for card-absent transactions). - Merchant category codes (MCC) to indicate type of business. - Identification approach: - Combine card issuer country of issuance with merchant outlet country to identify international (digitally ordered) transactions. - MCCs can help infer product types and may aid in separating digitally delivered services from goods. - Examples using card payment data: - Israel (Box 3.19): - Israeli CBS obtained monthly/quarterly credit card data from three major companies covering from 2012 onwards; data reported approximately two weeks after quarter end. - Data separate expenditures by Israeli tourists abroad and expenditures by Israeli residents cleared through foreign websites. - Data classified by MCC and limited to households (business cards excluded); only card-not-present transactions considered. - Spain (Box 3.20, OECD–BBVA collaboration): - Analysis limited to 2015 transactions. - Total number of online transactions recorded: 45.8 million in 2015, with a “total transaction value of several billion euros” across business and private customers. - Sample analysed: close to 60 per cent of the total transaction value; excluded business customers but accounted for over 96 per cent of all online transactions of private customers. - About 50 per cent of these transactions were outward‑bound, to a total of 115 countries. - Cross‑border payments from Spain were highly concentrated: Great Britain, Ireland and the Netherlands alone accounted for about 85 per cent of transactions involving foreign merchants. - Note: country‑specific legislation prevented identification of certain destination countries (e.g., transactions to merchants in Germany were not contained in the data). - Limitations and challenges: - Digital ordering is defined by the order being placed over computer networks, not by payment method; alternative payment methods (cash on delivery, wire transfer) can cause underestimation where cards are less prevalent. - Card-absent transactions can also occur for phone or in-person orders where card details are provided, causing potential misclassification. - Filtering out corporate/business cards is possible, but personal cards may be used for business expenditures (employee-paid, employer-reimbursed), creating noise. - Merchant outlet country in card data may reflect corporate principal place of business rather than actual seller location: - Visa rules require merchants to use principal place of business as merchant outlet location; corporate groups may have a single merchant location for all subsidiaries. - This can cause transactions to be recorded as domestic when the seller is actually abroad (or vice versa), especially for multinational headquarters and DIPs. - Transactions routed through DIPs can result in entire transactions being labelled international even when only a fee for intermediation should be recorded as cross-border. - Possible mitigation: identify payments to popular DIPs for separate treatment; differentiate domestic-currency DIP payments for services from foreign-currency payments. - Merchant category codes may not accurately reflect multi-product merchants. - Data access constraints: - Card payment data are held by private entities; access for statistical purposes requires agreements, data protection arrangements, potential payment, negotiation, or legislation. - Legislation in third countries can limit availability (example: Germany in Box 3.20). - Ongoing access is crucial; one-off historical extracts (e.g., Spain limited to 2015) limit time-series analysis. - In some countries, central banks publish card statistics (example: daily Credit and Debit Card Statistics by the Central Bank of Ireland). - Preconditions for effective use: - A favourable institutional environment, understanding of payment processes and nomenclatures, and widespread use of payment cards are key prerequisites. - Conclusion on card data utility: - If issues can be managed and access secured, card payment data can offer a relatively straightforward means to estimate household expenditure on digitally ordered purchases and households’ digitally ordered imports. ### Estimating the overlap between digitally ordered and digitally delivered trade - Definition: digital trade is “all trade that is digitally ordered and/or digitally delivered.” - Conceptual points: - Meeting either criterion (digitally ordered or digitally delivered) is sufficient to qualify a trade transaction as digital trade. - Many digitally ordered services are also digitally delivered (examples: media streaming subscriptions, many consumer telecommunications subscriptions, medical appointments booked online, digital intermediation services). - Such transactions are recorded within both digitally ordered and digitally delivered trade. - Statistical implication: - Totals for digitally ordered trade and digitally delivered trade cannot simply be summed to obtain total digital trade because that would double count transactions that are both digitally ordered and digitally delivered. - Measures of imports and exports of services that are both digitally ordered and digitally delivered should be compiled and entered separately to avoid double counting. *Source: Handbook on Measuring Digital Trade — Box 3.16 and adjacent boxes and text (Chapter 3 — Digitally Ordered Trade).* ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### Measuring the overlap: digitally ordered vs digitally delivered - Avoid double counting: when transactions are both digitally ordered and digitally delivered, the value reported under digitally delivered services (reporting template item 4) is subtracted when calculating total digital trade. - Need for sources that identify both total digitally ordered trade and the subset that is digitally delivered. - Country examples: - United Kingdom Digital Economy Survey: breaks down businesses’ e-commerce sales abroad and purchases from abroad by product (goods, digitally delivered services, other services); the digitally delivered services component yields an estimate of services that are both digitally ordered and digitally delivered. - Canada (business ICT usage survey, 2021): "62 per cent of private sector firms’ gross sales conducted over the internet were sales of physical goods, 18 per cent were digitally delivered services, and 20 per cent were other services" (Statistics Canada, 2022d). Although not specifically international, this can inform estimates of conceptual overlap between digitally ordered exports and digitally delivered exports. - Household and individual ICT usage surveys can collect: - amounts spent on digitally ordered and digitally delivered services, - online earnings from providing digitally delivered services. - Card payment data merchant category codes can identify digitally delivered services (example: Visa codes for “Cable, Satellite and Other Pay Television/Radio/Streaming Services” and codes for products “delivered in an electronic format” including books, films, digital artwork/images, music, games and applications (Visa, 2021)). - Combining multiple sources is especially useful to estimate the subset of digitally ordered trade that is also digitally delivered: - household expenditure surveys can indicate household spending on digitally delivered services and thus the share of households’ international e-commerce expenditures that are digitally delivered; - business registers or other business surveys can identify firms producing digitally delivered services and treat their e-commerce sales as digitally delivered. - Transparency: sources, methods, and assumptions used to measure or estimate services that are digitally ordered and digitally delivered should be clearly communicated to aid user understanding and international comparisons. ### Data sources and their roles - Business ICT surveys: - Can measure digitally ordered exports and imports by businesses, covering both goods and services. - Recommended to collect total business e-commerce sales and purchases and breakdown into domestic e-commerce and digitally ordered exports and imports, with product-level information (goods, digitally delivered services, other services). - Household ICT surveys and international travel surveys: - Can measure household e-commerce spending and online earnings; should collect product-level information and ask about digitally ordered accommodation and other travel-related expenditures. - Business core surveys and MNE surveys: - Core business surveys may offer larger coverage/sample sizes; MNE surveys can capture a significant portion of trade flows but cover only a subset of firms. - Hybrid strategies recommended: collect representative totals via core business surveys and breakdown details from ICT or other business sources. - Customs records: - Main source for merchandise trade; identifying digitally ordered shipments can yield integrated statistics on digitally ordered goods. - Limitations: excludes goods below customs thresholds (de minimis) and may require additional fields/processes to identify e-commerce shipments. - VAT returns: - Readymade source of business sales revenues but normally cover sales (exports) only and VAT-registered businesses; require methods to identify businesses selling online. - Card payments data: - Potential to estimate total digitally ordered household expenditures; can use merchant location and merchant category codes to identify international transactions and product types. - Limitations: proxy for digital payment not necessarily digital ordering; covers purchases (imports) only; may exclude non-card payments; product breakdown may be inexact; access may require negotiation, payment, or legislation. - Low-value parcel / de minimis trade: - Many low-value international goods shipments originate from digital ordering; administrative data from customs, postal and express courier agencies, and tax data can provide bases for estimation when coverage is high. - Integration imperative: - Information from different sources should be integrated to derive digitally ordered trade estimates representative of all institutional units; record and communicate coverage in terms of concepts, firm sizes, industries, etc., to facilitate correct use and international comparisons. ### Table 3.2 summary — strengths and limitations (by source) - Business ICT surveys - Strengths: measure exports and imports by businesses; cover goods and services; can measure conceptual overlap (reporting template item 4); designed for representative results; flexible to introduce new questions; integrate measurement with e-commerce. - Limitations: cover transactions involving businesses only; primarily designed to collect technology use rather than monetary amounts; may not cover all industries/firm sizes; integration challenges with other economic statistics. - “Core” business surveys - Strengths: integrated with macroeconomic statistics; may offer broader industry/firm size coverage and larger samples. - Limitations: cover businesses only; harder to add new questions than ICT/MNE surveys. - MNE surveys - Strengths: MNEs can underpin a considerable portion of digitally ordered trade; can collect goods and services trade. - Limitations: cover only a subset of businesses; may be challenging to exclude intra-economy affiliate transactions. - Household ICT surveys - Strengths: capture households as buyers and sellers; cover goods and services; can measure conceptual overlap and both imports/exports; designed for representative results and compatible with business surveys. - Limitations: cover households only; households may struggle to report amounts and identify international transactions. - Surveys of ICT usage in government units and NPISHs - Strengths: government units/NPISHs can be e-commerce buyers and sellers; cover goods and services and both imports/exports; designed for representative results and compatible with other surveys. - Limitations: such surveys are not widely implemented and typically have not measured spending or income from e-commerce/digitally ordered trade. - Customs records - Strengths: main source for merchandise trade statistics; identifying digitally ordered shipments can yield integrated statistics. - Limitations: goods only; excludes goods below customs thresholds (de minimis); requires implementation of data fields/processes to identify e-commerce shipments. - VAT returns data - Strengths: readymade data on business sales revenues. - Limitations: normally sales (exports) only; cover VAT-registered businesses only; requires methods to identify online sellers and may over-cover offline sales. - Card payments data - Strengths: cards are primary household payment means for online purchases; merchant location can identify international transactions; merchant category codes can enable product breakdown. - Limitations: purchases (imports) only; proxy for payment rather than ordering; not all card-not-present transactions are digitally ordered; excludes non-card payments; may mainly cover households but include corporate/personal-for-business payments; location info can be distorted by corporate structures; product breakdown inexact; access constraints. ### Key findings and illustrations - Importance of distinguishing digitally ordered and digitally delivered to avoid double counting when compiling total digital trade. - Practical country evidence: - Canada (2021): "62 per cent" goods, "18 per cent" digitally delivered services, "20 per cent" other services of private sector firms’ internet gross sales. - Merchant category coding (example from Visa, 2021) can identify electronic-format products and streaming services for measurement of digitally delivered services. ### Recommendations (numbered as in the chapter) 1. Assess which institutional sectors (businesses, households, government units, NPISHs) engage in statistically meaningful digital trade and prioritize measurement; prioritize business transactions in the absence of contrary evidence. 2. Use business ICT surveys to measure digitally ordered exports and imports by businesses; collect total business e-commerce sales and purchases and breakdowns into domestic e-commerce and digitally ordered exports/imports; collect product information (goods, digitally delivered services, other services); identify services digitally ordered and digitally delivered for aggregation of total digital trade. 3. Ensure business ICT survey coverage and estimation methods are sufficient to derive representative digitally ordered trade estimates for all businesses; follow relevant international guidelines (UNCTAD Manual for the Production of Statistics on the Digital Economy; OECD Model Survey on ICT Usage by Businesses). 4. Explore mainstreaming relevant questions into core business surveys used for structural business statistics and/or international trade in services surveys; use MNE surveys; consider hybrid strategies combining representative totals from core surveys with breakdowns from ICT/business sources. 5. Where manually typed emails are included, measure their value separately from e-commerce channels; if not possible, ask respondents whether they received orders or made purchases via email to indicate prevalence and potential scale. 6. Design business survey questionnaires with guidance to manage double reporting risks (e.g., online marketplace operators reporting only fees/commissions; industries where e-commerce concepts are less straightforward like financial services). 7. Seek to measure digitally ordered trade involving households as buyers and sellers; at minimum, measure total household e-commerce spending and earnings (domestic and international) to gauge their scale relative to business e-commerce; ideally collect product-level purchase/sale information. 8. Include questions in household and/or international travel surveys on expenditures for accommodation and other travel components that were digitally ordered; ask non-resident visitors similar questions; household income surveys should ask about short-term accommodation services supplied by residents that were ordered through DIPs to provide an upper limit for exports of accommodation by households. 9. Ensure household ICT survey coverage and estimation methods are sufficient to produce representative digitally ordered trade estimates for all households; follow international guidelines (ITU Manual for measuring ICT access and use by households and individuals (ITU, 2020) and The OECD Model Survey on ICT Access and Usage by Households and Individuals (OECD, 2015b)). 10. Encourage implementation of the WCO Framework of Standards on e-commerce and work closely with customs authorities to ensure customs reporting systems identify e-commerce shipments and reflect statistical needs. 11. Estimate many low-value international goods shipments (parcel trade), including those below de minimis thresholds, as they often result from digital ordering; use administrative customs data, postal and express courier agency data, and tax data where coverage is adequate. 12. Explore card payment data as a cost-effective way to estimate household digitally ordered expenditures; address challenges in identifying the international component and product types. 13. Integrate information from different sources to derive digitally ordered trade estimates representative of all institutional units; record and communicate sources and coverage in terms of concepts, firm sizes, industries, etc., to enable correct use and international comparisons. *CHAPTER 3 — DIGITALLY ORDERED TRADE, Handbook on Measuring Digital Trade.* ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### CHAPTER 3 — DIGITALLY ORDERED TRADE ### 4. Digitally delivered trade — Introduction - Definition: “All international trade transactions that are delivered remotely over computer networks”. - Scope: - Delivered remotely over the internet (including via mobile devices) and via private networks (e.g., via an extranet). - Only services are considered deliverable digitally under this Handbook’s convention. - Delivery can occur over a longer period and may include remote interpersonal interactions (e.g., video calls, manually typed emails, voice calls, fax messages, cloud networks). - Classification rule: - For transactions undertaken over multiple interactions or continuously, classification as digitally delivered or not should reflect delivery on an accrual basis according to how the service contract is fulfilled within the statistical period. - Overlap with digitally ordered trade: - Many digitally delivered services are also digitally ordered (notably downloadable and streamed products: software, music, video, e-books). - Not all digitally delivered services are digitally ordered (e.g., large-scale B2B procurements negotiated in person but delivered digitally). - To avoid double counting, the reporting template on digital trade (see Chapter 2, Table 2.1) includes a separate item for digitally ordered and digitally delivered services, which is to be subtracted when calculating the total. ### 4.2 Sources for measuring digitally delivered trade - Parties and intermediaries: - Transactions involve a buyer and a seller from any institutional sector: businesses, households, government bodies, NPISHs. - Digital intermediation platforms (DIPs) can act as intermediaries and facilitate delivery (see Chapter 5 for DIP measurement). - No single source covers the whole economy; multiple sources must be combined. - Key data sources and roles (as mapped in Figure 4.1): - ITS Surveys: natural starting point for businesses; effective for measuring digitally delivered trade by businesses (UN et al., 2010b). Excludes Mode 2 consumption while travelling (note 1). - Business ICT surveys: useful for identifying exporters/importers that make digital deliveries; sample design may be less suited to trade flow measures (note 2). - Household surveys: can capture household imports (streaming, etc.) but households may have difficulty delineating international transactions (note 3). - Travel surveys: cover digitally deliverable services consumed abroad (Mode 2) (note 4). - International Transaction Reporting System (ITRS): useful for large enterprises and transactions above reporting thresholds; minimum transaction thresholds may limit coverage (note 5). - VAT data: covers services imports subject to VAT (note 6). - Recommended compilation approach: - Combine ITS trade values with detail from ICT or business surveys to estimate the portion of trade flows that are digitally delivered. - First measure digitally deliverable services (Section 4.3), then apply expert judgment and business survey measures (Section 4.4) to estimate actual digital delivery. - Supporting references and tools: - Figure 4.1 maps potential sources by institutional sector and trade direction, with section references for further detail. - Table 4.6 summarizes strengths and limitations of each data source. ### 4.3 Digitally deliverable services — identification and guidance - Concept: - Digitally deliverable services = services which, at the time of writing, can be delivered through computer networks (most often the internet). - Some services may be technically deliverable but still delivered physically; delivery method in practice depends on availability of technology to supplier and consumer. - Some services may be a mixture of digital and non-digital delivery when traded internationally. - Rationale: - Many services are traded internationally only because digital delivery bridges physical distance (e.g., distance learning). - Table 4.1: Digitally deliverable services in EBOPS 2010 - Lists services recorded as digitally deliverable when supplied cross‑border (Mode 1) and those consumed abroad (Mode 2 recorded within Travel/SD). - Included EBOPS 2010 components (examples from Table 4.1): - Insurance and pension services (SF6) - Financial services (SG7) - Charges for the use of intellectual property n.i.e. (SH8) - Telecommunications, computer and information services (SI9) - Research and development services (SJ110.1) - Professional and management consulting services (SJ210.2) - Architectural, engineering, scientific and other technical services (SJ3110.3.1) - Trade-related services (SJ3410.3.4) - Other business services n.i.e. (SJ3510.3.5) - Audio-visual and related services (SK111.1) - Health services (SK2111.2.1) - Education services (SK2211.2.2) - Heritage and recreational services (SK2311.2.3) - Notes: - SDMX‑BOP DSD codes referenced for EBOPS 2010 items. - For Mode 2, recommended to use Travel breakdown by product rather than by purpose (see Box 4.3). - An expanded version of Table 4.1 is available in Annex C. - Additional inclusions and exclusions: - Additional categories included given potential digital delivery: “health services” and “heritage and recreational services”. - Digital intermediation platform fees are included in “Trade-related services”. - Inherently digital services: software downloads, cloud computing, streaming media, online gaming (including virtual items/real estate), digital communication services, datasets — mainly recorded in “Audio-visual and related services” and “Telecommunications, computer and information services”. - Insurance and financial services transactions generally in scope for digitally delivered trade (even if not determined by deliverability). - Services with digital outputs (reports, designs, blueprints) are commonly digitally delivered (examples: accounting, management consultancy, R&D, architectural services — including “digital twin” delivery). - Services mainly physical in nature (manufacturing services, repair and maintenance, construction) are generally excluded because digitally delivered component values are expected to be small relative to overall item values. - NFTs (non-fungible tokens) are not included in the EBOPS 2010 based list of digitally deliverable services in Table 4.1. - Measurement recommendation: - First target the services items listed in Table 4.1 when measuring digitally delivered trade. - This keeps scope aligned with cross-border (Mode 1) services supply per UN et al., 2010a. - Acknowledges that most digitally delivered trade occurs through cross-border supply (Mode 1) but also recognizes Mode 2 consumption abroad (example: a telehealth consultation while travelling). - Countries are encouraged to assess relevance for further services categories and to pursue research/experimentation to inform potential extensions in future Handbook editions. ### Box 4.1 — Mobile money and digital trade (key points) - Definition: Mobile money = digital medium of exchange and store of value usually offered by a mobile network operator (MNO); does not require a traditional bank account. - Uses: - Commonly used for personal remittances and can be used to pay for cross‑border goods and services. - Typical actors in a mobile money transaction: buyer/donor, seller/recipient, buyer’s MNO, seller’s MNO (possibly the same), local mobile money agents converting cash/credit across countries; MNOs and agents charge fees. - Accounting treatment: - Use of mobile money for payment does not determine whether the underlying product is digitally ordered or delivered. - Fee payments to MNOs and between MNOs are recorded as international trade in services if the buyer/donor is resident in a different economy than the MNO, or if one MNO pays fees to another MNO in a different economy when the two parties are resident in different economies. - These fees are payments for financial services provided by telecom companies and should be recorded as digitally delivered services trade. - Empirical note: - Pilot surveys to record these services were conducted in 2017‑18 in Uganda, Botswana and the Philippines (Bank of Uganda, 2018). *Source: IMF, OECD, UNCTAD and WTO.* ### section 4.3.2) are synonymous with measuring ### hmdtea - section 4.3.2) are synonymous with measuring ### Role of product detail in delineating digitally deliverable services - Collecting sufficient product detail is a prerequisite for accurately delineating digitally deliverable services within wider international trade in services statistics. - Aggregated trade in services classes can be used to indicate services that can be digitally delivered, but they do not cover all services in Table 4.1 and should be built upon. - Trade flows in digitally deliverable services provide an upper-bound estimate of digitally delivered trade; they will be greater than (or equal to) the total value of services actually digitally delivered. - UNCTAD (with Costa Rica, India and Thailand) found a high proportion of exports of many digitally deliverable services appear to be actually digitally delivered, but in India up to one-fifth of digitally deliverable exports were still delivered by non-digital means. ### Box 4.2 — 3D printing and digital trade (key points) - 3D printing (additive manufacturing) involves loading a 3D design file into a machine that prints 3D structures in layers using plastic, resin, concrete, metal or other materials. - The act of 3D printing is inherently physical; printing services are physically, not digitally, delivered, and the resulting objects are physical goods. - Digital delivery plays an important role via the design files which are easily traded internationally through the internet; online services offering design files for paid download should be recorded as trade in digitally delivered services. - Source attribution in box: IMF, OECD, UNCTAD and WTO. ### Databases and cross-country comparisons - International trade in services databases useful for cross-country comparisons and global trends: - WTO-UNCTAD common dataset on international trade in services (covering 200 economies). - OECD EBOPS 2010 Trade in Services by Partner Economy database. - Figure references: Figure 4.2 (US trade in digitally deliverable services, 2010-21) and Figure 4.3 (Digitally deliverable services exports by region, 2015-21). ### Towards measures of digitally delivered services (concepts and implications) - Identifying a service as digitally deliverable does not mean it is always delivered digitally when traded. - Trade in digitally deliverable services is an upper-bound estimate; further steps are needed to measure services actually digitally delivered. - Mode 1 (cross-border) transactions imply physical distance between buyer and seller; for services that can be digitally delivered, digital technology will generally bridge that distance. - Mode 1 delivery accounts for the majority of trade in digitally delivered services; measures of the portion of digitally deliverable services imported and exported by Mode 1 offer reasonable estimates for the bulk of digitally delivered trade. - Two main approaches to delineate the portion of digitally deliverable services supplied via Mode 1: - Estimates based on research and expert judgement shares. - Measures collected through business surveys. ### Estimates based on expert judgement - A first step: apply shares based on expert judgement to products identified in Table 4.1. - Assessments should ideally be made on a country-by-country basis to account for: state of digitalization, prevalence of digital technologies and skills, trading partners, etc., and regularly updated to reflect technological advancements. - For countries without resources for bespoke exercises, standard allocation shares established internationally can be used (see Box 4.6). - TiSMoS and consolidated models: - The MSITS 2010 framework suggests a simplified allocation approach by likely mode(s) used to supply services for each balance of payments item. - The WTO Trade in Services by Mode of Supply (TiSMoS) methodology operationalizes MSITS 2010 simplified allocation and enabled production of the first global dataset of trade in services by mode of supply in 2019. - Eurostat and the WTO developed a consolidated Eurostat‑WTO model (included in Annex D). - In 2023, the WTO produced a global dataset on digitally delivered services based on the Eurostat‑WTO model and adjusted allocation shares for the pandemic and survey results. - Global composition (2022 estimates) of Mode 1 exports of digitally delivered services: - “Other business services” (including “research and development services” and “professional services”, such as “legal and management consulting services”) — 40 per cent. - “Computer services” — 20 per cent. - “Financial services” — 16 per cent. - “Charges for the use of intellectual property n.i.e.” — 12 per cent. - “Personal, cultural, and recreational services” (including music and video streaming services) — 3 per cent. - Additional sector shares shown in Figure 4.5: Insurance and pension services - 5%; Telecommunications services - 3%; Information Services - 1%. ### Measures collected through business surveys (ITS surveys) - International trade in services (ITS) surveys covering businesses provide the best means for obtaining more precise estimates of digitally delivered services trade. - Supplemental questions should be asked only for services that can be delivered digitally (not necessary for services digitally delivered by nature, e.g., streaming media subscriptions). - To reduce reporting burden, supplemental questions could be asked less frequently than routine ITS questionnaires or targeted at main exporters/importers of relevant products. - BEA (United States) and ONS (United Kingdom) have developed methods providing estimates of digitally delivered trade using the same survey questions used to delineate services trade by modes of supply; Statistics Canada used a similar approach. - Remote delivery concept: - Remote delivery includes delivery by post as well as digital delivery; in BEA, ONS and Statistics Canada, remotely delivered but non-digital services are judged to be marginal, so cross-border remotely delivered services provide meaningful estimates of digitally delivered trade. - Survey design practice: - BEA, ONS and Statistics Canada asked respondents to estimate, for various product classes, the share of trade that was remotely delivered by selecting from pre-defined percentage ranges (with option “unknown” in the UK case). - BEA requested Mode 1 delivery information only for service sectors expected not to be supplied exclusively through Mode 1 to lower respondent burden. - ONS and Statistics Canada did not restrict the enquiry, leading to some Mode 1 transactions reported for further services items (e.g., maintenance and repair services, construction). - Example: BEA’s survey question format collected exports/imports presence and, for each “Yes”, respondents checked one percentage range: - Less than 25% - 25-49% - 50-74% - 75-89% - 90-99% - 100% - Respondents indicated whether the information provided was based on “Accounting records” or “Recall/general knowledge of operations”. ### Box 4.3 — Measuring digitally deliverable services consumed abroad (Mode 2) (key points) - The “travel” component in the balance of payments covers goods and services consumed by non-residents in the economy they visit (Mode 2). - Some Mode 2 services are digitally delivered (e.g., telecommunication services via e‑sim, digital guides, remote medical consultations, e‑learning). - Identifying digitally delivered services in travel is challenging; household surveys compiling travel transactions could be amended to ask whether a service was digitally delivered, but respondents may not know the counterpart country. - A first step: adopt EBOPS “alternative presentation for travel” broken down by product rather than by purpose (Annex I), components include: - 4a.1 Goods - 4a.2 Local transport services - 4a.3 Accommodation services - 4a.4 Food‑serving services - 4a.5 Other services (Of which: 4a.5.1 Health services, 4a.5.2 Education services) - Presentation by product helps identify Mode 2 services transactions; goods, local transport, accommodation and food‑serving services cannot be digitally delivered; category 4a.5 could encompass some digitally delivered services and provide an upper bound estimate, although likely only a small share of “other services” are digitally delivered. - Source attribution in box: IMF, OECD, UNCTAD and WTO and Bank of Italy. *Source: Handbook on Measuring Digital Trade — CHAPTER 4 — DIGITALLY DELIVERED TRADE (content provided from the supplied PDF excerpt).* ### 1. This question applies to the following 13 transaction types, which are expected to have Mode 1 transactions, which ma ### 1. This question applies to the following 13 transaction types, which are expected to have Mode 1 transactions, which may be digitally delivered: “accounting, auditing, and bookkeeping services”; “advertising services”; “other computer services”; “education services”; “architectural services”; “engineering services”; “surveying, cartography, certification, and technical inspection services”; “legal services”; “market research services”; “public opinion and polling services”; “other management, consulting, and public relations services”; “provision of customized and non-customized research and development services”; and “other research and development services”. ### Digitally deliverable services and measured Mode 1 shares - The Handbook notes high levels of digital delivery for services identified in Table 4.1, but measured shares of remote delivery for imports and exports are markedly below 100 per cent. - Eurostat-WTO model shares (finalised in 2021) reflect accelerated digitalisation during the COVID-19 pandemic; observed shares in any given economy and industry may deviate from the model averages. - Importance of supplementing expert-judgement model shares with survey questions on trade in services to obtain more accurate, country-specific measures of digitally delivered exports and imports. ### Eurostat‑WTO model versus ITS survey-based Mode 1 comparisons (Table 4.3) - Eurostat-WTO model (Annex D) compared with international trade in services survey results (exports and imports), percentage: - Accounting, auditing, and bookkeeping services: Eurostat-WTO model 75; Survey-based Exports 51; Imports 66 - Advertising, market research, and public opinion polling services: Eurostat-WTO model 75; Survey-based Exports 78; Imports 70 - Computer services: Eurostat-WTO model 75; Survey-based Exports 80; Imports 56 - Architectural and engineering services: Eurostat-WTO model 75; Survey-based Exports 61; Imports 53 - Education services: Eurostat-WTO model 75; Survey-based Exports 37; Imports 32 - Legal services: Eurostat-WTO model 75; Survey-based Exports 80; Imports 91 - Management consulting and public relations services: Eurostat-WTO model 75; Survey-based Exports 77; Imports 68 - Research and development services: Eurostat-WTO model 90; Survey-based Exports 59; Imports 81 ### National experiences measuring digitally delivered transactions via ITS surveys - United States (BEA) - For its 2022 Benchmark Survey, BEA refocused questions to target digitally delivered trade by excluding supply by postal service or private delivery; expanded instructions explicitly include services provided via extranet, fax and video conference. - BEA requests a point value for the share of trade digitally delivered (rather than a percentage range) in 2022. - Number of service types surveyed increased from 13 to 18; additional service types: “news agency services”; “installation, alteration, and training services”; “operational leasing”; “trade‑related services”; and “health services”. - United Kingdom (ONS) - ONS included response category “unknown” plus the 6 percentage ranges used by BEA. - Pilot of 100 businesses in September 2018 informed addition of questions to 2018 annual ITS survey of 5,000 businesses. - ONS did not restrict inquiry to services judged ex ante to be remotely deliverable; respondents reported Mode 1 delivery for products not typically considered remotely deliverable (e.g., manufacturing services, maintenance and repair, construction), indicating need for further research into reported perceptions of remote delivery. - Canada (Statistics Canada) - Moved from simplified allocation to direct measurement in the Annual Survey of International Transactions in Commercial Services by adding a single question on remote delivery of services exports. - Respondents indicate share of each service product delivered remotely using six standard percentage ranges. - Findings: computer services producers digitally delivered 93 per cent of their computer services exports; education services had 60 per cent of export value arising from Canadian educators relocating abroad (pre‑COVID); COVID‑19 generated a significant rise in digitally delivered education services in 2020 and 2021. - By classifying commercial services exports at enterprise level, digitally delivered exports grew 25 per cent from 2019 to 2020, while non-digitally delivered commercial services exports registered a slight decline. - Costa Rica (UNCTAD collaboration) - Targeted 220 enterprises exporting services on the UNCTAD list of “potentially ICT‑enabled services”; 171 responses, 119 reported exporting services delivered digitally. - Grossed up to population of 1,391 firms using Central Bank and administrative records. - Results for 2021: - 90 per cent of those firms digitally delivered services internationally in 2021. - Of all exports of the targeted services, 94 per cent were digitally delivered in 2021. - Digitally delivered targeted services amounted to 51 per cent of total services exports and 20 per cent of total exports. - Digitally delivered exports contributed 7.2 per cent to the gross domestic product (GDP) of Costa Rica in 2021. - Over three‑quarters of firms exporting ICT‑enabled services were foreign‑owned; parents predominantly from the European Union or United States. - UNCTAD pilot countries showed over 80 per cent of exports of the digitally deliverable services covered were actually digitally delivered. ### ONS Eurostat‑WTO vs ITS survey comparison (Table 4.4) — selected entries - Table 4.4 reports Eurostat-WTO model (Annex D) and ITS survey (2020, 2018) percentages for exports and imports across many service types. Selected figures: - Insurance and pension services: Eurostat-WTO model 100; ITS survey (2020) Exports 91; ITS survey (2018) Exports 84; ITS survey (2020) Imports 66; ITS survey (2018) Imports 71 - Financial services: Eurostat-WTO model 100; ITS survey (2020) Exports 95; ITS survey (2018) Exports 89; ITS survey (2020) Imports 88; ITS survey (2018) Imports 79 - Telecommunications, computer and information services: Eurostat-WTO model 80; ITS survey (2020) Exports 83; ITS survey (2018) Exports 85; ITS survey (2020) Imports 86; ITS survey (2018) Imports 85 - Other business services: Eurostat-WTO model 80; ITS survey (2020) Exports 89; ITS survey (2018) Exports 65; ITS survey (2020) Imports 78; ITS survey (2018) Imports 65 - Travel: Eurostat-WTO model 0; ITS survey (2020) Exports 0; ITS survey (2018) Exports 0; ITS survey (2020) Imports 0; ITS survey (2018) Imports 0 - Transportation: Eurostat-WTO model 90; ITS survey (2020) Exports 65; ITS survey (2018) Exports 65; ITS survey (2020) Imports 80; ITS survey (2018) Imports 80 - Government goods and services n.i.e.: Eurostat-WTO model 10; ITS survey (2020) Exports 75; ITS survey (2018) Exports 75; ITS survey (2020) Imports 75; ITS survey (2018) Imports 75 ### Survey design, implementation guidance, and caveats - Recommendations and cautions when using ITS surveys to measure digitally delivered trade: - Avoid approaches that ask respondents only to identify their main mode of supply for a given service; such approaches do not yield sufficient information to derive estimates of digitally delivered trade. - Provide clear instructions and train field/telephone agents to assist respondents who have difficulty breaking down trade across modes of supply. - Implement checks to detect potential misreporting for follow-up when reported digital/remote delivery shares are significantly different from simplified model suggestions. - Note that some services not listed as digitally deliverable in Table 4.1 may be reported as remotely deliverable (examples: manufacturing, maintenance and repair, construction). - Implication 1: Care and guidance are needed to ensure respondents correctly record transactions in EBOPS 2010 classes and as remotely delivered or not; outsourcing to a third party should not be recorded by the principal as digital remote delivery. - Implication 2: The list of services considered digitally deliverable may need expansion in future, but the Handbook currently recommends remaining consistent with Table 4.1 and Annex C. - When compiling statistics from ITS surveys: - Propensity for digital delivery measured from sampled firms will be grossed up to the target population according to compilers’ sampling methodology; estimates may be determined by responses from a relatively small number of firms. - Digitally delivered trade by larger firms is likely well measured; imputations for unsurveyed small- and medium-sized firms and uncertainty about behavioural correlation across similar firms affect robustness. - Communicate caveats and limitations of data and methods to users and stakeholders. - Consider developing a flag in business registers/survey population lists identifying “digitally-oriented firms” to guide sampling for trade in services surveys. - When primary data are not available for a unit, consider using prior-period responses with adjustment or information from other enterprises in the same enterprise group. ### Compiling digitally delivered transactions using business ICT surveys - Business ICT surveys can gather information on firms’ use of digital means of delivery where adding questions to ITS surveys is infeasible. - Business ICT surveys often addressed to IT departments; recommend stating that respondents may need input from sales/accounting colleagues for questions on digital delivery. - Suggested questions for business ICT surveys: 1. [During the reporting period] did your business use digital means to deliver services products to customers? yes/no. 2. If yes: what was the revenue from sales of these digitally delivered services? % or $. 3. Breakdown of revenue from sales of digitally delivered services to customers located in: - a. Own country % or $ - b. Other countries % or $ 4. Of the revenue from all sales of digitally delivered services (question 2), what was the value of sales where the service was also digitally ordered? % or $. - 4a. Of the revenue from sales of digitally delivered services to customers abroad (question 3a), what was the value of sales where the service was also digitally ordered? % or $. - Explanatory text should define “digital means” as delivery through video calls, manually typed emails, voice calls, fax messages or any other digital communication devices, as well as through cloud networks. *Source: Handbook on Measuring Digital Trade, Chapter 4 — Digitally Delivered Trade.* ### 4.5 Other sources ### 4.5 Other sources ### 4.5.1 Compiling digitally delivered transactions using ITRS data - The International Transaction Reporting System (ITRS) can provide scope to estimate digitally delivered services, especially for transactions involving large enterprises that predominantly provide digitally delivered services. - The experience of Brazil (Box 4.11) shows this approach is feasible and can derive separate estimates of intra-firm digitally delivered trade. - The ITRS can be used to select the largest international traders for each EBOPS item; direct interaction or a small survey of those enterprises can then be used to estimate digitally delivered services. - A similar approach can begin from a business register or other administrative source to select the largest enterprises and interview them. - The ITRS may be used to identify payments to and from non-resident Digital Intermediary Platforms (DIPs). Care is needed to ensure that, when the DIP intermediates transactions between buyers and sellers both resident in the compiling economy, only the amount relating to intermediation services provided by the DIP (and not the value of the services being intermediated) is recorded in digitally delivered trade. Box 4.11: Digitally delivered services in Brazil - The Central Bank of Brazil (Banco Central do Brazil (BCB)) traces international trade in services flows using the ITRS. The Brazilian ITRS covers all foreign exchange settlements between resident businesses and non‑residents. - The Brazilian ITRS has more than 50 different codes to identify types of services transactions, allowing allocation in the balance of payments with a good level of detail. - Every transaction is registered (i.e., no threshold is in place) and has a national fiscal registration number identifying the resident party; for the non‑resident party, the name is provided. - BCB contacted several of the largest enterprises operating in Brazil to better understand their business models and decide on an appropriate allocation of transactions to digital trade categories. - Virtually all foreign multinationals operating in Brazil that deliver services digitally to residents also have international transactions with their foreign parent companies; these international transactions are the focus for measurement of digitally delivered trade. - Example: one large MNE has a Brazilian subsidiary that sells online advertising space to customers in Brazil, is physically present in Brazil and employs over 100 staff (software developers and sales assistants). The subsidiary purchases online advertisement services from its parent company and provides them to local customers in Brazil. ### 4.5.2 Compiling digitally delivered transactions using tax administration data VAT data (general) - Some countries collect value-added tax (VAT) on services digitally delivered into their country by foreign actors. With simplifying assumptions—most notably about the products sold by each firm—this can provide a source of data on digitally delivered trade. - Box 4.12 (Argentina) and section 4.2.2(ii) provide examples. Box 4.12: VAT data in Argentina - Law No. 27430/2017 stipulates that the 21 per cent VAT rate also applies to digital services provided by non‑residents to residents. Resident financial intermediaries that act as agents in the collection of this tax are asked to provide information on these transactions. - The fiscal authority data cannot be disaggregated by product detail, so additional information is requested directly from the intermediaries. - A detailed concordance between the firms covered and the services they supply was developed by assuming non‑resident firms export products related to their main activity and allocating them to EBOPS 2010 categories. Examples of assignments: - Credit rating services and other financial services → explicitly charged and other financial services (EBOPS 2010 component 7.1). - Services of messages, calls and video calls provided through internet protocol (e.g., Skype or Viber) → telecommunications services (9.1). - Computer services (9.2): - manufacturers/distributors of antivirus software (e.g., Symantec or Panda) → 9.2.1 computer programmes; - applications for webpage creation/design (e.g., WordPress) → 9.2.2 other computer services; - web hosting, servers or domains (e.g., Bluehost) → 9.2.2 other computer services; - platforms for downloads of videogames or other computer software (e.g., Sega or PlayStation Network) → 9.2.1 computer programmes. - Information services (9.3.2): web hosting for information, images, video or other content (e.g., Yahoo or Truvalia); subscription services to digitized newspapers/magazines. - Accounting and related services (e.g., PWC) → accounting, auditing, bookkeeping, and tax consulting services (10.2.1.2). - Business and management consulting and public relations services (10.2.1.3): consulting through videoconferences or other digitized means (e.g., Neelus). - Instagram, Facebook and Twitter → advertising services, market research and public opinion surveys (10.2.2). - Intermediation platforms connecting buyers and sellers of business services → other business services not elsewhere included (10.3.5) (e.g., Habitissimo); employment services that charge premium services (e.g., DGNet, LinkedIn) → 10.3.5.1. - Audio‑visual and related services (11.1.1): streaming services (e.g., Spotify and Netflix). - Remote education services (e.g., OpenEnglish) → other personal, cultural and recreational services (11.2.2). - Sporting and gambling services (e.g., Betsson, Bwin) → heritage and recreational services (11.2.3). - “Free” dating platforms with advertising revenue (e.g., Tinder or Badoo) → other personal services (11.2.4). - For multi‑product firms (e.g., Google Play), anecdotal evidence was used to split products between EBOPS categories. - For firms providing both digital services and goods/non‑digital services, Argentina adopted a threshold of US$ 10; above this, transactions are assumed not to relate to digitally delivered services. - Because payment is mainly through credit cards, it was assumed the main resident sector involved was the household sector. - Caveats: - Intermediation services for platforms intermediating goods cannot be estimated with this method; their commission, in theory, is captured in goods statistics (valued at Cost, Insurance and Freight (C.I.F.)). - The US$ 10 threshold for mixed firms may be an issue, though anecdotal evidence suggests it is not currently significant. VAT data — the European Union One Stop Shop (OSS) - Compilers in the European Union have used VAT Mini One Stop Shop (MOSS) scheme data to progress on cross-border, business-to-consumer (B2C) services transactions. The services covered under MOSS are digitally delivered and include website hosting, supply of software, access to databases, downloading apps or music, online gaming and distance teaching. - Under MOSS, companies that supply certain cross-border services to individuals and other entities not liable for VAT in the EU may file VAT returns in a single member state. The VAT is collected by the tax authorities in that member state and redistributed quarterly to other member states. - The tax authorities in each member state receive from counterpart countries: the name of the company, its VAT number, the country of registration, whether the company is an EU resident, and the value of the sales in the reporting period. These data can be used to estimate household sector spending on cross-border digital services if made available to national statistics offices or other compilers. - MOSS captures many smaller household transactions but has challenges: - Companies are not obliged to use MOSS; bigger suppliers may file VAT returns through other means. - Data may include cross-border payments by other non-taxable entities (e.g., government or education service providers), requiring care to avoid double counting. - Other challenges include timing and country detail; granular company-level data access helps address these. - The new OSS (One Stop Shop) scheme, in place since 2021, extends MOSS and covers B2C cross-border transactions including “distance” sales of goods and electronic services. - Early OSS experience shows goods need to be separated from services; compilers should ensure no overlaps with data from other sources (e.g., VAT Information Exchange System (VIES), survey data, counterpart export data). Box 4.13: The use of VAT Mini One Stop Shop (MOSS) data in Denmark - Before MOSS, Statistics Denmark estimated supply of digital services to private consumers using various sources for five categories (streaming, apps, gambling, games and other services). - Introduction of MOSS data led to significant revisions: in 2015, imports of computer services were revised upwards to DKK 2.9 billion from DKK 0.4 billion. - MOSS data showed imports by private individuals accounted for 6 per cent of all imported computer services and almost 30 per cent of audio‑visual services. - Statistics Denmark expects the move from MOSS to OSS/IOSS will have only a small impact. Box 4.14: The use of “One Stop Shop” (OSS) data in Hungary - Exports: - A comparison for 14 enterprises identified in both OSS and the International Trade in Services Survey (ITSS) showed those 14 enterprises accounted for only 14 per cent of the total exports included in OSS. - ITSS figures were generally higher than OSS, as ITSS covers a wider scope of services. - More than 62 per cent of OSS exports are represented by 12 enterprises, of which only two are included in ITSS. - OSS data accounted for 0.5 per cent of total Hungarian EU services exports in 2021. - Imports: - OSS can be used to estimate digitally delivered services by households, which are not covered by ITSS. - OSS imports for 2021 were higher than the import grossing‑up in ITSS for relevant EBOPS 2010 codes, implying household digital services may be underestimated in ITSS (billing differences may also explain the discrepancy). - As a share of total EU services imports, OSS data amounted to 3.6 per cent. - Future plans: - Use OSS to refine selection of data providers dealing with digital trade and refine grossing‑up methods in exports. - Plan to link non‑resident enterprises with EBOPS 2010 codes and use OSS imports to refine grossing‑up in ITSS by estimating digital services used by households. ### Table 4.5 (selected statistics from Argentina): Imports of digitally delivered services paid by households in the third quarter of 2022 - Audio‑visual and related services (11.1.1): 123,865,939 US$ - Computer software (9.2.1): 33,660,666 US$ - Advertising services (10.2.2): 21,090,180 US$ - Other computer services (9.2.2): 17,836,474 US$ - Employment services (10.3.3.1): 4,341,608 US$ - Heritage and recreational services (11.2.3): 3,169,463 US$ - Telecommunication services (9.1): 979,120 US$ - Education services (11.2.2): 946,995 US$ - Other business services n.i.e. (10.3.5): 839,533 US$ - Other personal services (11.2.4): 726,280 US$ - Business and management consulting (10.2.1.3): 305,883 US$ - Financial services (7.1): 161,627 US$ - Information services (9.3.2): 118,069 US$ - Accounting services (10.2.1.2): 4,288 US$ - TOTAL: 208,046,125 US$ ### 4.5.3 Compiling digitally delivered transactions with household surveys - Households are active consumers of digitally delivered services (streaming music/video, online gaming) and telecommunications services that enable digital delivery; households may also act as producers of digitally delivered services (e.g., selling video or audio content online). - Household consumption (imports) and sales (exports) of digitally delivered services are unlikely to be well captured in trade statistics, which traditionally rely on enterprise surveys or ITRS. - Household surveys can gather information on household spending and earnings online, including digitally deliverable products, but households face challenges calculating expenditures and identifying transactions as international versus domestic. - It may be possible to build on experiences measuring online spending by products and online earnings (e.g., Canada, Box 4.16) to identify the international dimension. ### 4.6 Recommendations - There is no single approach that offers easy and complete measurement of all dimensions of digitally delivered exports and imports. Based on the examples, the following recommendations are identified: 1. Defining digitally delivered services: - For a service to be digitally delivered, it is a prerequisite that it is digitally deliverable. - Efforts should target the digitally deliverable services identified in Table 4.1 and detailed further in Annex C. 2. Using expert judgement: - In the absence of appropriate data sources (e.g., survey questions), estimates can be derived by applying expert judgement shares of the portion of each service product delivered by cross-border (Mode 1) supply. - These shares can be based on various sources, including observations from countries with similar characteristics (notably, with a similar level of digitalization), but must be applied at a sufficiently detailed degree of product disaggregation. 3. Compiling digitally delivered services based on ITS surveys: - The collection of data on digitally delivered trade through ITS surveys is recommended as a priority. - ITS surveys should collect sufficient product detail (and sub-product detail as necessary) to allow digitally deliverable services to be distinguished from other services as a basis for statistical compilation. - There are synergies with collecting information on Mode 1 supply of services (among other modes of supply to trade services). *Source: Handbook on Measuring Digital Trade, Chapter 4 — Digitally Delivered Trade.* ### Box 4.15: Estimating household expenditures on digital services in Ireland ### Box 4.15: Estimating household expenditures on digital services in Ireland ### Overview of the Irish exercise (2022) - In 2022 the Central Statistics Office (CSO) of Ireland combined administrative sources with publicly available data to compile estimates of the expenditure on digital services by households. - The CSO used articles, studies and reports from private companies and researchers to research the provision of online services by firms that are not already in the VAT OSS dataset for inclusion in the estimation. ### Table 4.6: Expenditure on digital services by households in Ireland (2020, millions of euros) - Table title as given: TABLE 4.6: EXPENDITURE ON DIGITAL SERVICES BY HOUSEHOLDS IN IRELAND - Column headings as given: By product and region/country of the seller, 2020, millions of euros - Region/Country and values as presented in the source: - Ireland252111 - Europe (excluding Ireland)2631181284260 - North America11344 - Rest of the world006 - Not allocated251641 - Source line in table: Source: Ireland Central Statistics Office (2022). ### Related examples and household survey evidence - Canada (CIUS) experience: - The Canadian Internet Use Survey (CIUS) collected information on individuals’ expenditure on various digital services and asked explicitly: "How much did you spend on the following digital services during the past 12 months?" with items including music or video downloads or streaming subscriptions, E‑books, audio books or podcast books, Online newspapers or magazines, Online gambling, Online gaming, gaming applications, game downloads or in‑game purchases, Any additional digital services ordered over the internet, and Total. - A similar module in the 2018 and 2020 editions found: - average expenditure per individual on digital services was CA$ 568 in 2020, - an increase of almost 40 per cent compared to 2018, - in 2020, spending on digital services comprised around 17 per cent of average total online expenditure on all goods and services. ### Methodological sources and complementary approaches highlighted - The chapter identifies multiple complementary sources and approaches to measure household consumption/imports of digitally delivered services: - International Transaction Reporting System (ITRS) - Administrative data such as VAT records (including regimes to collect VAT from non-resident digital services providers) - Household surveys (adding questions on digitally delivered products and travel/border surveys for Mode 2) - ICT/business surveys and ITS (International Trade in Services) surveys (including adding questions on digital/remote delivery) - Key methodological notes and guidance: - For digitally deliverable services products that are readily available, aggregate to give a measure of “digitally deliverable services trade” as an upper-bound estimate of digitally delivered trade. - For digitally deliverable products not likely to be 100 per cent digitally delivered, add questions on digital/remote delivery to ITS surveys; the UNCTAD model questionnaire (UNCTAD, 2021a) is a recommended starting point. - ICT surveys can be used to obtain the percentage of exports of services that were digitally delivered and the share of digitally ordered products; these data should be combined with international trade in services statistics to derive product and geography breakdowns. - Administrative VAT data can be a very useful source for household imports of digitally delivered services where VAT is collected from non-resident providers. - ITRS can be useful at the total economy level but requires investigation of individual companies to derive product breakdowns. ### Strengths and limitations of sources (summary from Table 4.7) - Digitally deliverable services (compiled from ITS survey data) - Strength: Can be compiled using product detail commonly available in existing services trade statistics. - Limitation: Digitally deliverable ≠ digital delivery; does not cover Mode 2. - Digitally delivered services (Mode 1) – estimated by expert judgement shares - Strength: Leverages modes of supply framework; initial estimates without extra data collection. - Limitations: Does not cover Mode 2; Mode 1 may include delivery by post; standard shares may not reflect country specifics. - Digitally delivered services (Mode 1) – measured through ITS survey questions - Strength: Measures digitally delivered trade without separate Mode 1 questions. - Limitation: Implementing remote delivery questions requires resources and adds respondent burden. - Business ICT surveys - Strength: More flexibility to add new questions; can measure overlap between digitally ordered and digitally delivered trade. - Limitation: No real-world examples of business ICT surveys including digital delivery modules; combining with ITS figures may be challenging without a central business register. - International Transaction Reporting System (ITRS) - Strengths: Ready-made source; supplemental information can be included with low respondent burden; suited to identifying large companies. - Limitations: Potential misclassifications when banks report; transactions recorded when payments are made not at time of output; counterpart country may not correspond to service origin; requires quality checks and reporter training. - VAT data - Strength: Ready-made source when VAT is collected from non-resident digital services providers. - Limitations: Only available when VAT is collected; only covers businesses subject to VAT; VAT reporting may not separate digitally and physically delivered services. - Household surveys - Strength: Can cover imports and exports of digitally delivered services by households not captured by ITS or ITRS. - Limitation: Households may have difficulty distinguishing international transactions from domestic. ### Practical compilation guidance and communication - Integrate information from different sources to derive estimates representative of all institutional units in the whole economy. - Record and communicate the sources used and coverage of digitally ordered trade estimates in terms of concepts, firm sizes, industries, etc., to enable correct user interpretation and facilitate international comparisons. - Lack of estimates for some items (e.g., digital intermediation services provided by DIPs, Mode 2 consumption abroad) should not preclude aggregation of digitally deliverable services trade based on available data or estimation of digitally delivered Mode 1 trade. *Source: Ireland Central Statistics Office (2022); Handbook on Measuring Digital Trade, Chapter 4.* ### Chapter 2, Figure 2.1) and in the reporting template ### CHAPTER 5 — DIGITAL INTERMEDIATION PLATFORMS (DIPs) ### Definition and examples of DIPs - Digital intermediation platforms (DIPs) are defined as: - “Online intermediation services that facilitate transactions between multiple buyers and multiple sellers in exchange for a fee, without the online intermediation unit taking economic ownership of the goods or rendering the services that are being sold (intermediated).” - Examples of DIPs include: - marketplace platforms that bring together buyers and sellers to trade goods and services, e.g., platforms facilitating short-term accommodation; - platforms facilitating ride hailing, similar to taxi services; - platforms facilitating sharing of household assets, such as car-sharing; and - platforms that intermediate electronic content (without taking economic ownership of the intellectual property products they distribute), such as app stores. - All institutional sectors can use DIPs. Non-financial corporations and the household sector in particular use DIPs both as buyers and as sellers. - The Handbook assumes transactions undertaken via a DIP are digitally ordered; in some cases services facilitated through DIPs may also be digitally delivered. ### Classification of DIPs and intermediation services (Section 5.2) - Work on industry and product classification is ongoing: - TT-ISIC (Task Team on ISIC) coordinates ISIC Rev.4 and recommends DIPs be classified in the industry producing the products they intermediate (e.g., wholesale or retail trade for goods). - NAICS 2022 follows similar principles: platforms intermediating the sale of goods are classified indistinguishably with e-tailers and bricks-and-mortar retailers; platforms intermediating services are classified in the industry of the service they intermediate. - Product classification for digital intermediation services is under discussion in the context of the Central Product Classification revision. - BPM6 guidance recommends classifying digital intermediation services under trade-related services (commissions on goods and service transactions payable to merchants, brokers, dealers, auctioneers and commission agents) (BPM6, paragraph 10.158). ### Accounting principles for DIP transactions (Section 5.3) - Transactions facilitated by DIPs involve at least three actors: buyer (consumer), seller (producer), and the DIP. - When at least one actor is resident in a different economy, transactions must be recorded in international accounts. - DIPs do not take ownership of goods nor render the intermediated services; their match-making role is treated as an arranger (BPM6, paragraph 3.10). - Main transaction (provision of good or service) is recorded in the accounts of seller/producer and buyer/consumer; the DIP’s accounts show only the fee charged for facilitation services. - Recording depends on the residence of the three actors; the supporting matching service provided by a non-resident DIP is a cross-border provision of a digitally ordered and digitally delivered intermediation service. - The Handbook aligns with SNA and BPM guidance and the first edition (2019) of the Handbook. ### Unpacking DIP transactions and fee treatment (Section 5.3.1) - Distinguish supply of goods/services (seller-buyer) from provision of intermediation services (DIP to seller and buyer). - DIPs are remunerated through fees from buyer, seller, or both; fees may be explicitly invoiced or implicit (not itemized). - Table 5.1 scenarios (preserve compiler knowledge perspective): - Explicit: fees paid by buyer and/or seller are known → record fees paid from buyer and/or seller to DIP. - Implicit (known payer, unknown amount): estimate fees paid from buyer and/or seller to DIP. - Implicit (unknown payer and amount): estimate total fee and show total paid by the seller to DIP. - If fees are implicit, the difference between what the buyer pays and what the seller receives can be assumed to reflect the value of digital intermediation services, or estimate fees based on DIP reports operating in the compiling economy. - When it is not possible to establish who pays the fee, assume intermediation fees are entirely incurred by the seller. - Fees may be paid at transaction time, earlier, later, or through regular payments; record on an accrual basis. - Promotional terms, rebates, or zero fees do not negate that a digital intermediation service was provided; discounts may result in negative fees in practice. ### Numerical example and reporting-template recording (Box 5.1 and Table 5.2) - Example assumptions: - Buyer pays 100 for a good, of which 12 is a fee paid to the DIP. - DIP charges a fee of 8 to the seller. - Buyer, seller and DIP are resident in different economies. - Payment is routed through the DIP. - Recommended recording steps: 1. Buyer makes payment of 100 to the DIP. DIP itemizes buyer’s payment for intermediation services as 12. - Buyer country records 12 as imports of digital intermediation services and 88 as imports of goods. 2. Seller liable for fee of 8, itemized by the DIP; seller country records 8 as imports of digital intermediation services. 3. DIP forwards payment for product to seller but subtracts the 8 owed by seller; seller receives 80 from the DIP. 4. For correct net trade of seller, statistical reporting must show a payment of 88 from buyer to seller (80 received by seller plus 8 charged to seller). - Table 5.2 summary (Buyer country / Seller country / DIP country; items and numeric entries preserved as in source): - Item: Total digital trade 2+3 minus 4 → Buyer country Exports 100 / Imports 88 / Seller country Exports 8 / Imports 8 / DIP country Exports 20 / Imports (blank) - Digitally ordered trade 2.1+2.2 → Buyer: 100 / 88 / 8 / 12+8 - 2.1 Goods → 88 / 80+8 - 2.1.a of which: via DIPs → 88 / 80+8 - 2.2 Services → 12 / 8 / 12+8 - 2.2.a of which: via DIPs → (as above) - Digitally delivered trade 3 → 12 / 8 / 12+8 - 3.a of which: via DIPs → (as above) - Services digitally ordered and digitally delivered 4 → 12 / 8 / 12+8 - 4.a of which: digital intermediation services → 12 / 8 / 12+8 - Net income from trade → -100 / 80 / 20 - Notes from example: - Total imports = 100 + 8 = 108. - Total exports = 88 + 20 = 108. - Total trade in goods = 88. - Total trade in services = 20. - System is balanced and each country’s net trade is correct. - For intermediation of services, recording is similar but main transaction is recorded in items 2.2 and 2.2.a rather than 2.1 and 2.1a, and also in items 3, 3.a and 4 if digitally delivered. ### Recording across residence scenarios (Table 5.3) - Table 5.3 prescribes recording of transactions under different residence and payment scenarios: - If seller pays the intermediation fee OR it is unknown who pays: - Examples include import by country B from country A with none (domestic) intermediation service; import by country B from country A and import by country A from country B; and others depending on which parties are resident where. - If buyer pays the intermediation fee: - Examples include import by country B from country A and import by country B from country A (intermediation services also imported by country B), etc. - If both seller and buyer pay the intermediation fee: - Examples include import by country B from country A and import by country B (of part of the intermediation services) from country A with remainder domestic, and other permutations where imports of intermediation services are split between relevant countries. - Compilers should apply Table 5.3 guidance to attribute product and intermediation-service treatments depending on residence of seller, DIP and buyer. ### Additional compilation guidance and complexities (Sections 5.4–5.5 overview) - Measurement and compilation of DIP activity remain challenging and exploratory. - Section 5.2 discusses classification; Section 5.3 describes accounting principles; Section 5.4 offers guidance on measurement of DIPs established in the compiling economy and experiences compiling imports of digital intermediation services; Section 5.5 summarizes main recommendations. - Transactions facilitated by DIPs can be more complex than the basic three-party example: - Additional roles may exist (e.g., transport/delivery, warehousing). - The DIP may directly provide further services (fee covers intermediation plus further services) or purchase further services from non-residents (may give rise to international trade in services). - Transport or warehousing may be part of another intermediation arrangement; payment may be split among DIP, seller, and other service providers. - Compilers should carefully examine arrangements and apply appropriate recording. *Source: IMF, OECD, UNCTAD and WTO (Handbook on Measuring Digital Trade — Chapter 5: DIGITAL INTERMEDIATION PLATFORMS (DIPs)).* ### 5.4 Measuring DIP ### 5.4 Measuring DIP ### Overview and stocktaking findings - There has been limited experience of measuring the activities of DIPs in many countries, including developing economies. - Large DIPs provide intermediation services in multiple countries, and there are examples of DIPs having residency in developing countries. - Measurement of international trade transactions involving DIPs is a pressing issue for all economies. - An OECD–IMF stocktaking survey in 2018 (OECD, 2018c) found that few compilers can identify the amount of trade facilitated by DIPs (either domestic or foreign-owned), and fewer still can identify payments to non-resident DIPs. - DIPs resident in a given economy should be in the statistical business register, but they are often included under various industry headings and formal identification remains difficult. - Manual identification of the largest DIPs, based on the name of the business, was reported as a feasible method to facilitate compilation of statistics on goods and services traded via DIPs and on digital intermediation services. ### Identifying DIPs - Initial detection efforts were largely based on manual identification and focused on peer‑to‑peer online platforms. - A 2017 EU study identified nearly 500 peer‑to‑peer digital intermediation platforms active in Europe, of which 4 per cent had over 100,000 unique website/app visitors per day. - The UK Office for National Statistics (ONS) followed a similar manual and statistical learning approach in early work to identify and measure the sharing economy. ### Compiling information from DIPs (business reporting) - Business surveys can measure: - Prevalence of DIPs in the economy. - Fees received by DIPs from residents and non-residents. - Information on transacted products. - International trade in services (ITS) surveys are well placed to collect information on exports and imports of digital intermediation services; survey instructions should explicitly state that trade‑related services cover digital intermediation services and ideally identify them separately. - The US Bureau of Economic Analysis (BEA) applied this approach in its Benchmark Survey of Selected Services and Intellectual Property Transactions with Foreign Persons. - The BEA has expanded the survey so that if an enterprise self‑identifies as a DIP, the questionnaire requests information on income from intermediation fees and the service type under which those intermediation services are reported. - ITS surveys may not be well suited to collect detailed information on the transacted products; other business surveys or enterprise ICT usage surveys can be better placed for collecting values of goods and services intermediated by DIPs and their international shares (items 2.1.a and 2.2.a of the reporting template on digital trade). ### Challenges observed in MNE and business surveys - BEA experience (2019 Benchmark and subsequent work) highlighted challenges: - Lack of responses leading to incomplete coverage of digital intermediation activities. - Misinterpretation of the digital intermediation services question by reporters, including reporting sales where firms did not act as intermediaries or conflating fee‑based platforms with data‑ or advertising‑driven platforms. - The digital intermediation services question was the most challenging for reporters to interpret; BEA is researching methods to refine results by estimating values where coverage is incomplete and identifying over‑reported values. - Current quarterly and annual ITS surveys should capture cross‑border payments by enterprises to DIPs if accompanying information notes state that fees paid for digital intermediation services are recorded under trade‑related services. ### Enterprise ICT usage surveys as a preferred vehicle - Enterprise ICT usage surveys are modular and can collect detailed digital topics; they are suitable to gather: - Sale of goods via DIPs (and amount exported) - Sale of services via DIPs (and amount exported) - Purchase of goods via DIPs (and amount imported) - Purchase of services via DIPs (and amount imported) - Fees paid to DIPs (and portion that are imports) - The UK ONS Digital Economy Survey has asked enterprises to state amounts paid in fees to DIPs to sell goods and services; questions can be extended to total goods and services sold and export shares. ### Compiling information from DIP users (demand-side measurement) - Many DIPs operate in economies without a physical presence; fees paid to a non‑resident DIP constitute an import of digital intermediation services, making measurement especially challenging. - Challenges from buyer perspective include: - Households may not know the value of the fee, even if explicit. - Respondents may find it difficult to determine whether the transaction was with a non‑resident or resident DIP (e.g., perceived local intermediation if seller is resident or DIP uses a local domain name). - Some progress has been made measuring the value of underlying goods and services transacted via DIPs using business and household surveys. ### Business surveys for DIP users - Businesses are key DIP users as sellers and buyers and are more likely than households to know the residency of the DIP. - Business surveys, including ITS and enterprise ICT usage surveys, should: - Capture information on goods and services intermediated by DIPs. - Record fees paid by enterprises to DIPs and identify cross‑border transactions. - Provide data needed for items 2.1.a, 2.2.a and 4.a from the digital trade reporting template (Table 2.1, Chapter 2). - Mainstream business surveys could include questions on amounts sold or purchased via DIPs; coverage advantages may offset lower topical detail. ### Household surveys for DIP users - Household surveys can be used successfully in some countries to compile statistics on purchases intermediated by well‑known DIPs. - For estimating trade facilitated by DIPs and intermediation fees via household surveys, important information to collect includes: - Total value of goods and/or services intermediated. - Proportion of these goods and services transacted with non‑residents. - Transaction fee paid to the DIP. - Travel is a sector significantly transformed by DIPs; travel transactions are typically measured on the demand side via tourism expenditure surveys. - Examples: - Canada collects demand‑side information about DIP activity related to transport and accommodations in its Canadian Internet Use Survey (CIUS). - The 2018 OECD‑IMF Stocktaking Survey indicated several countries use tourism expenditure survey questions to measure travel booked through DIPs. - The Spanish National Statistics Institute (INE) established that in 2017, 68 per cent of outbound tourists booked accommodation (excluding hotels) using an online intermediation platform. - Italy used a similar approach on its border survey to collect information on frequency of using online tools for booking or buying travel‑related services. *Source: "5.4 Measuring DIP", hmdtea - 5.4 Measuring DIP (Handbook on Measuring Digital Trade, CHAPTER 5 — DIGITAL INTERMEDIATION PLATFORMS (DIPs)) — content provided from the source PDF.* ### Chapter 3, Box 3.2). ### Chapter 3, Box 3.2) ### Overview - Discussion focuses on measuring transactions intermediated by Digital Intermediation Platforms (DIPs) and related data collection challenges and approaches. - Emphasizes distinction between the value of goods and services intermediated and the intermediation fee; notes many household surveys collect only the value of the goods and services intermediated and not the fee nor the residency of the DIP. - Highlights multiple data sources and methods: household surveys, web scraping/big data, payment card data, app store transaction data, business surveys, targeted surveys of large global DIPs, and tourism statistics. ### Household survey evidence and issues - Household surveys in some countries collected the value of goods and services purchased via DIPs but: - Questions only include the value of goods and services being intermediated. - None of the survey questions asked about the fee paid by the household to the DIP, nor do the questions concern the residency of the DIP. - If the value of the fee is not collected, compilers may: - Estimate the fee as a percentage on the value of the goods and services intermediated based on other known examples (perhaps from DIPs in the reporting economy). - Use expert knowledge and apply information based on activities of a few companies, since DIP activity is often dominated by a few very large companies. - Example: France (Box 5.9) included questions in its household panel survey targeting domestic and outbound tourism; the Banque de France survey asks: - How was the booking made? Options: (1) phone, (2) internet / application, (3) face‑to‑face. - What type of operator was used? Options include (3) online intermediation platform and (4) aggregator / search engine. ### Alternative data sources and techniques - Web scraping / big data: - Approach: identify DIPs by scraping company websites and linking with the statistical business register; enrich business register using key words and expressions. - Netherlands developed a 2016 publication on the digital economy using such techniques (Oostrom et al., 2016). - Statistics Indonesia used web scraping for e‑commerce characteristics, Consumer Price Index weights, and occupancy rates for tourism statistics (see Box 5.10). - Challenges: legal issues, large data storage needs, data‑quality issues, differing platform structures requiring complex automatic navigation. - Recommendation from Statistics Indonesia: have a partnership framework with DIPs and a legal basis for web scraping. - Payment card data: - Considered by Belgium, Estonia, Finland, France, Israel, Latvia and Mexico (2018 OECD‑IMF Stocktaking Survey) to measure imports of digital intermediation services. - Caution: payment card data may indicate payment to a non‑resident DIP but further information/assumptions are needed to separate the intermediation fee from the goods/services intermediated; the intermediated good or service may or may not be imported and may or may not be collected from other sources. - Transactions in apps / app stores: - App stores act as DIPs; when individuals buy an app, part may be an intermediation fee to the app store. - Some countries may access third‑party provider data on app transactions; must understand whether data include the app store intermediation service, which may need to be imputed for import measurement. ### Targeted approach: surveying DIPs directly - Option: target large global DIPs with questionnaires requesting breakdowns of: - Value of goods and services intermediated. - Intermediation fees. - Permission to share country‑level information with other compilers (subject to data sharing agreements). - Benefits: - Would significantly improve coverage of DIPs and estimates of imports/exports of goods and services intermediated by DIPs, particularly for countries where DIPs are not established or compilers have data sourcing challenges. - Example: Eurostat publishes monthly experimental statistics on short‑term accommodation based on data provided by four international platforms under data exchange agreements (monetary information not included). ### Tourism examples and key statistics - Spain (INE) Residents Travel Survey collects accommodation booking mode with a specific category for digital platforms when accommodation is a rented holiday home or room in a private dwelling. Booking mode options include: - (1) Directly with the service provider through its web or app - (5) Through a specialized webpage (e.g., AirBnb, Homeaway, Booking.com, Homelidays, Niumba, Rentalia, Housetrip, Wimdu, Interhome, Friendly Rentals, etc.) only if Q1 was = s3 or 4 - Results for Spain (2021): - Domestic trips: for rented holiday homes booked through platforms: - 5.3 per cent of trips - 6.0 per cent of nights spent - 9.7 per cent of total expenditure - Outbound trips: for rented holiday homes booked through platforms: - 11.9 per cent of trips - 7.4 per cent of nights spent - 10.1 per cent of total expenditure - Booking channel shares (Figures 5.3 and 5.4): - Domestic (all domestic trips using any accommodation, 2021): Specialised website (platform) 49% ; Service provider no internet 26% ; Service provider internet 10% ; Real estate internet 2% ; Don't know 9% ; Real estate no internet 3% ; Included in tourism package 1%. - Outbound: Specialised website (platform) 77% ; Service provider internet 9% ; Don't know 4% ; Included in tourism package 2% ; Service provider no internet 7% ; Real estate no internet 0% ; Real estate internet 1%. ### Recommendations (chapter summary) - The chapter provides the following recommendations for recording and measuring DIP activity: 1. Recording DIP transactions: - Distinguish the supply of goods or services (seller–buyer transaction) from provision of intermediation services (DIP–seller/buyer transaction). - Explicit fees should be recorded as a payment to the DIP for intermediation services, from the buyer and/or the seller as appropriate. - Implicit fees need to be imputed (based on the difference between what the buyer pays and what the seller receives, or separately estimated from reports by DIPs). - When it is not possible to establish who pays the fee, assume intermediation fees are entirely incurred by the seller. - Digital intermediation services should be recorded in the balance of payments under trade‑related services. - Regardless of explicit or implicit fee treatment, the main transaction between buyer and seller should reflect either: - the full value the buyer pays less the fee paid by the buyer to the DIP (if any); or - the value of the good or service being intermediated plus the intermediation fee paid by the seller to the DIP. 2. Identifying and surveying DIPs in the compiling economy: gather information on prevalence of DIPs and use business surveys to measure intermediation services traded and underlying intermediated goods and services. 3. Measuring exports and imports of digital intermediation services by enterprises: ITS surveys should collect exports of intermediation services by resident DIPs and imports from non‑resident DIPs; enterprise ICT usage surveys can complement. 4. Measuring exports and imports of goods and services via DIPs by enterprises: enterprise ICT usage surveys (or other business surveys) should collect details on exports and imports intermediated by DIPs. 5. Measuring imports via DIPs and imports of digital intermediation services by households: household surveys should include questions on value of goods and services purchased via DIPs, separately identifying domestic and non‑domestic purchases, and the value of intermediation fees where known; at minimum include questions on well‑known DIPs. 6. Measuring DIPs transactions in the tourism sector: travel and/or tourism statistics compilers should gather information on the value of transport and accommodation services facilitated by DIPs and associated intermediation fees. 7. Targeting global DIPs directly: national and international statistics agencies should explore targeted surveys of large global DIPs with cross‑border data sharing arrangements. ### Strengths and limitations of data sources (summary) - ITS surveys: measure exports/imports of intermediation fees by country from resident/non‑resident DIPs; integrated with services trade statistics; need added questions for value of goods/services intermediated; limited scope to add questions due to response rate/respondent burden. - Business ICT surveys: flexible to include detailed questions on trade via DIPs and fees; may not cover all industries/firm sizes. - Core business surveys: can measure DIP‑facilitated sales/purchases with a new question; may have good coverage but limited scope to add questions. - MNE surveys: can capture significant portion of trade involving large DIPs; cover subset of businesses; require new questions. - Household ICT surveys: households can in principle report total amount paid; successful for well‑known DIPs; respondents may struggle to isolate DIP purchases, delineate intermediation fees, or identify residency of DIP/supplier. - Tourism surveys: easy integration of questions on accommodation and travel expenditures via DIPs; respondents may have difficulty delineating fees and identifying DIP residency. *Source: Handbook on Measuring Digital Trade — CHAPTER 5 — DIGITAL INTERMEDIATION PLATFORMS (DIPs).* ### CHAPTER 5 — DIGITAL INTERMEDIATION PLATFORMS (DIPs) ### CHAPTER 5 — DIGITAL INTERMEDIATION PLATFORMS (DIPs) ### Overview and objectives - China Customs prioritized measuring digitally ordered trade (cross-border e-commerce, CBEC) due to rapid growth and policy interest. - First measurement efforts started in 2014 with specific customs procedure codes and preliminary “CBEC statistics from customs control perspective”. - Since 2021, statistics combining customs records, enterprise surveys, and other sources have been published as “CBEC statistics from business perspective”. - Measurement objectives follow IMTS 2010 and four key principles: - Integration: integrated approach for scope, sources, processing methods, and release mechanism. - Reliability: robust survey process and sound processing methods. - Innovation: use of web-scraping to extract share of Chinese goods sold by overseas platforms. - Independence: CBEC statistics are a new statistical product included in external merchandise trade statistics with lower granularity and frequency than customs-control statistics. ### Data sources used - Customs administrative records using CBEC-specific procedure codes: - Codes: 9610 (cross-border direct purchases), 1210 (bonded cross-border purchase), 1239 (bonded cross-border purchase A), 9710 (cross-border e-commerce between enterprises), 9810 (overseas warehouse of cross-border e-commerce exports). - Based on codes 9610, 1210 and 1239, China Customs estimated CBEC transaction values of US$ 13.3 billion in 2017 and US$ 27.5 billion in 2019. - The five codes (9610, 1210, 1239, 9710 and 9810) are the primary sources for CBEC statistics. - Enterprise survey (conducted twice a year): - Respondents classified into four roles: a) third-party platforms; b) self-built/self-operated platforms; c) e-commerce enterprises providing CBEC-related services; d) e-commerce vendors on third-party platforms. - When one respondent holds multiple roles, all respective questionnaires are required; only platforms (categories a and b) are used to estimate total trade; others used for cross-validation and analysis. - CBEC enterprises of significant scale complete detailed questionnaires including origin and destination; questionnaire available in Annex E (source). - Financial reports of overseas platforms: - Gross sales, revenues and public information drawn biannually from financial reports of platforms such as Amazon, eBay, Shopee and Wish to calculate CBEC settlements. - Settlement value defined as amount paid by buyers to vendors for Chinese goods sold on these platforms; normally exceeds export (FOB) value. - Qualitative interviews: - Interviews with service providers to obtain information on logistics, payment and platform management costs for deduction from platform settlements when estimating FOB exports. - Other sources: - 2019 telephone survey (with Ministry of Commerce and State Postal Administration) of consignors of about 5,000 randomly selected outbound postal parcels in six cities: Harbin, Qingdao, Chengdu, Wuhan, Hangzhou and Guangzhou. This determined share of CBEC goods in outbound parcels; share for inbound parcels obtained through industry interviews with postal parcel operators. - Since 2019, CBEC parcels included in external merchandise trade statistics; statistical value derived by multiplying average price indicated on parcels by number of parcels to minimize impact of price outliers. - Proxy codes introduced where essential commodities lack corresponding HS codes. ### Coverage and classification - Scope consistent with WCO Framework of Standards on Cross-Border E-Commerce (WCO, 2018); primarily applies to B2C and C2C and encourages extension to B2B. - Transactions recorded as CBEC goods when: - Trade occurs between enterprises or individuals within and outside China’s customs territory; - Orders are placed through an internet platform; - Goods are delivered through cross-border logistics channels. - CBEC statistics compiled and presented as a subset of total merchandise trade statistics. ### Compilation methodology — imports - Process for CBEC imports: - Step 1: Retrieve customs administrative records with codes 9610, 1210 and 1239. - Step 2: Extract survey results of import values on goods not explicitly declared as CBEC to customs. - Aggregate customs records as Vi1 (“value import 1”) and survey results as Vi2 (“value import 2”). - Divide Vi2 by coverage rate of sampled enterprises (Rsi) obtained from enterprise survey to reduce omissions. Rsi is the estimated share of transactions by sampled enterprises in total CBEC goods. - Total import value Vi derived using the formula: - Vi = Vi1 + Vi2 / Rsi - Results verified by comparing shares of main platforms in major markets with previous measurements in total trade of consumer goods. ### Compilation methodology — exports - Measuring exports ordered via domestic platforms (Ve1): - Export values reported in platform questionnaires are aggregated as Ve1. - Domestic platforms include third-party and self-run/self-built platforms (examples: Alibaba, Jindong, DHgate, Youzan and Shein). - Measuring exports ordered via overseas platforms (Ve2): - Use sources in Section 1.2 for overseas platforms (Amazon, eBay, Shopee, Wish). - Example: Amazon export estimation process: - Extract revenue turnover of Amazon online stores and third-party seller services from Amazon financial reports. - Apply Amazon commission, storage fee, distribution fee and other expenses (obtained by interviewing Chinese sellers) to convert revenue turnover into a settlement value for Amazon third-party and self-operated goods (Ea, the ratio of Chinese goods sold via Amazon). - Analyse seller address information to estimate sales of Chinese goods via Amazon (Ra, value of the goods themselves, excluding services), provided by a third party, then derive settlement value of Chinese goods on Amazon (S1’). - S1’ = Ea * Ra - Convert S1’ into FOB-based exports (S1) by applying charges expressed as a percentage (C), where charges include platform fees, logistics costs, customs clearance fees, taxes and other CBEC fees. - S1 = S1’ * (1‑C) ### Practical implementation details - New customs procedures introduced: - 2014: code 9610 introduced for cross-border direct purchases to streamline customs treatment of digitally ordered B2C goods with simplified clearance and aggregated declarations; CBEC enterprises or agents required to submit e-orders, e-payment vouchers and e-waybills with a declaration. - Late 2014 and 2016: codes 1210 and 1239 introduced for bonded cross-border purchases used in bonded logistics customs centres or bonded zones. - June 2020: codes 9710 and 9810 introduced to extend CBEC customs control from B2C to B2B. - Aggregation and reporting practices: - By the 15th of every month, CBEC enterprises or agents summarize prior month’s simplified declarations and transform them into a single formal declaration under code 9610, consolidated by exporter, mode of transport, country of consignment, country of last-known destination, commodity code and port of export to facilitate tax refunds and identify CBEC records in the database. - Statistical treatment for parcel values: average price per parcel × number of parcels to derive statistical value and reduce impact of declaration errors. *Source: CHAPTER 5 — DIGITAL INTERMEDIATION PLATFORMS (DIPs), Handbook on Measuring Digital Trade (case study on Measuring digitally ordered merchandise trade in China).* ### CHAPTER 6 — CASE STUDIES ### CHAPTER 6 — CASE STUDIES ### Data integration and verification (China Customs CBEC methodology) - CBEC total export sold on major overseas platforms (Ve2) is obtained by summarizing Chinese exports ordered by different platforms (as S1, S2... Sn). - Total CBEC export (Ve) calculation: - Ve1 = exports ordered from domestic platforms - Ve2 = exports ordered from overseas platforms - Rse = sample enterprises coverage rate (obtained by enterprise surveys) - Ve = (Ve1+Ve2)/Rse - Data verification and validation practices: - Business interviews adjust market proportions of major platforms and compare domestic vs overseas platform ratios. - If interview results and estimations differ greatly, China Customs investigates omissions or errors. - Ratios of CBEC imports and exports to total consumer products imports and exports and their percentage changes are checked. - CBEC statistics use survey information, customs administrative records, and parcel data (including personal effects) for preliminary and annual estimations. ### Preliminary aggregate results (CBEC, China, 2022) - Aggregate 2022 totals and growth: - China’s total international CBEC trade in 2022: US$ 309 billion, up by 3.8 per cent year-on-year. - CBEC exports in 2022: US$ 230 billion, increased by 6.7 per cent. - CBEC imports in 2022: US$ 79 billion, shrank by 3.8 per cent. - Main markets and product composition: - Main export markets (in order of market size): the United States, the United Kingdom, Malaysia, France, Germany, Japan, Spain and Russia. - Main import sources: the United States, Japan and the Republic of Korea. - Consumer goods share of CBEC exports: 93 per cent (major items: clothing, shoes and bags; textiles and other household items; electronic products). - Consumer goods share of CBEC imports: 98 per cent (major items: perfumes, make-up and other personal care and toiletry items; health products; maternal and children’s products). ### Dissemination and breakdown of CBEC statistics - Dissemination timetable and revisions: - CBEC statistics are compiled and released annually, with preliminary estimations available every quarter. - Quarterly preliminary data are not revised. - CBEC imports and exports data are produced with a four-month lag due to source availability; merchandise trade statistics are usually released within one month. - A quarterly estimation mechanism was established in 2021 to improve timeliness. - Preliminary statistics lack breakdowns; annual estimation updated with survey data and other sources. - Breakdown categories and classification approach: - CBEC statistics broken down by trading partner (country/region), end use, import and export modes, and domestic locations (provincial level). - Questionnaires classify CBEC goods into consumer goods and inputs of production; consumer goods are divided into 10 sub-groups for imports and 10 sub-groups for exports. - Consumer goods classification (imports) — exact sub-groups: - Food and drinks; - Milk powder; - Nappies; - Perfumes, make-up and other personal care and toiletry items; - Medicines and health products like food supplements; - Mobile phones, computers and other digital products, household appliances and peripheral products; - Textiles and other household items; - Clothing, shoes and bags; - Toys, infant and children’s products other than milk powders; - Other consumer goods. - Consumer goods classification (exports) — exact sub-groups: - Clothing, shoes and bags; - Mobile phones, computers and other digital products, household appliances and peripheral products; - Textiles and other household items; - Jewellery, clocks and watches, glasses; - Toys, infant and child products other than milk powders; - Articles for entertainment and sports; - Perfumes, make-up and other personal care and toiletry items; - Utensils for gardening and home use; - Peripheral products of automobiles; - Other consumer goods. - Historical CBEC import/export figures (Table 6.1: imports and exports of CBEC in China, 2019-22): - 2019: Total US$ 187, Export US$ 116, Import US$ 71.0, Year-on-year (%) Total 24.9, Export 6.1 - 2020: Total US$ 234, Export US$ 157, Import US$ 77.0, Year-on-year (%) Total 25.2, Export 35.5, Import 8.7 - 2021: Total US$ 297, Export US$ 215, Import US$ 82.7, Year-on-year (%) Total 27.0, Export 37.4, Import 6.1 - 2022: Total US$ 309, Export US$ 230, Import US$ 79, Year-on-year (%) Total 3.8, Export 6.7, Import -3.8 ### Conclusions: lessons learned (China Customs CBEC) - Key success factors: - Sound cooperation among stakeholders (commerce, postal services, statistics, Chamber of Commerce). - Incorporation of survey responses by major domestic CBEC platforms reduces omission and duplication; platforms are core data providers for digitally ordered imports and exports. - Harmonization and adjustment of platform-sourced retail prices (which include platform fees, logistics fees, taxes and other post-declaration fees) with customs FOB/CIF valuations enable integration with merchandise trade statistics. - Standards and valuation: - CBEC goods included after physical movement across borders in line with IMTS 2010 and domestic Regulations on Customs Statistics. - Export valuation on a FOB basis; import valuation on a CIF basis. ### Challenges (China Customs CBEC) - Collaboration with overseas data producers is more difficult than with domestic platforms. - Granularity and frequency limitations: - CBEC statistics are of low frequency and low granularity relative to conventional merchandise trade statistics derived from customs administrative records. - Parameter reliability and quality assurance: - Insufficient ways to check reliability of parameters used in estimation. - Need for sustainable and cost-effective parameter updating and maintenance. ### Future steps (China Customs CBEC) - Measures to improve data quality and coverage: - Applying big data technology to improve data quality and compilation methodology for totals and breakdowns. - Holding more interviews with independent “station construction service providers” and enterprises providing logistics and payment services to enhance survey coverage. - Differentiating survey questionnaires by respondent type to reduce response burden and obtain consistent information. - Strengthening partnerships with data-holders. --- ### Case study 2 — Towards a better measurement of digitally delivered trade: China’s experience and prospects #### Introduction and background (MOFCOM) - MOFCOM responsibilities: promoting foreign trade, participating in WTO e-commerce negotiations, advancing digital trade development. - 2006 policy: “Thousand-Hundred-Ten project” to promote exports of three categories of ICT services regarded as digitally deliverable: information technology outsourcing (ITO), business process outsourcing (BPO) and knowledge process outsourcing (KPO). - Outcome metrics cited: over 67,000 enterprises engaging in these activities; over 10 million jobs created; exports reaching 200 economies worldwide. - Data systems: - Investigation System on Service Outsourcing introduced in 2007. - Online Monitoring System on Service Outsourcing launched in 2009. - Investigation system reviewed every three years; MOFCOM conducts countrywide personnel training. #### Compilation process and data sources (MOFCOM) - Concepts and classification: - International services outsourcing: services provided by Chinese enterprises to overseas clients. - Categories: ITO, BPO, KPO (Table 6.2 lists sub-categories and delivery methods; ITO: All/almost all digitally delivered; BPO/KPO: Partially digitally delivered). - Statistical principles: - Completeness: statistical scope, indicator list, measurement approach, data dissemination set via Investigation System. - Timeliness: registration forms and system upgraded every three years; new contracts and completed service deliveries updated and disseminated on a monthly basis. - Accuracy: three-layer review mechanism for validation and anomaly detection. - Main data source and coverage: - Online Monitoring System on Service Outsourcing (launched 2009) requires enterprise registration and monthly contract details. - By 2021, over 67,000 enterprises registered; about 10,000 enterprises participate in the survey each year. - Surveys conducted by MOFCOM centrally and by local commercial departments; validated data summarized after governmental reviews. - Questionnaire collects enterprise information, service details, contract clauses, contract values and implementation status, personnel and international certifications (questionnaire available in Annex F). #### Preliminary results (international service outsourcing, China, 2019-22) - Aggregate values and growth: - 2022 international service outsourcing: US$ 136.8 billion (from US$ 96.9 billion in 2019); average annual growth of 12 per cent. - International ITO accounts for over 40 per cent of total and rose from US$ 42.7 billion to US$ 56 billion (average annual growth of 9.5 per cent). - Table 6.3 (International service outsourcing, 2019-22) — exact figures: - 2019: International service outsourcing Value (US$ billion) 96.9; International ITO Value (US$ billion) 42.7; International BPO Value (US$ billion) 17.5; International KPO Value (US$ billion) 36.7 - 2020: International service outsourcing 105.8; International ITO 46.3 (YoY change 7.1); International BPO -2 (YoY change -42.3); International KPO 42.3 (YoY change 15) - 2021: International service outsourcing 130.3; International ITO 55.0 (YoY change 19.8); International BPO 19.8 (YoY change 16.5); International KPO 55.5 (YoY change 31) - 2022: International service outsourcing 136.8; International ITO 56.1 (YoY change 0.2); International BPO 22.4 (YoY change 13.1); International KPO 58.3 (YoY change 7.0) - International ITO sub-category breakdown (Table 6.4) — exact values and YoY changes: - International ITO total: - 2019 Value (US$ billion) 42.7, YoY change 6 - 2020 Value 46.3, YoY change 55 - 2021 Value 55.0, YoY change 19 - 2022 Value 56.1, YoY change 0.2 - IT research and development services: - 2019 Value (US$ billion) 34.4, YoY change -0.1 - 2020 Value 37.8, YoY change 104 - 2021 Value 45.4, YoY change 204 - 2022 Value 44.9, YoY change -1.0 - IT operation and maintenance services: - 2019 Value (US$ billion) 7.1, YoY change 36 - 2020 Value 6.7, YoY change 78 - 2021 Value 8.7, YoY change 149 - 2022 Value 9.6, YoY change 9.4 - IT application development services: - 2019 Value (US$ billion) 1.2, YoY change 155 - 2020 Value 0.8, YoY change -290.9 - 2021 Value 0.9, YoY change 91 - 2022 Value 1.6, YoY change 8.2 - Cross-validation with balance of payments aggregates (Table 6.5): - International ITO (US$ billion) vs Telecommunications, computer and information services exports (US$ billion) and Ratio: - 2019: International ITO 42.7; Telecommunications, computer and information services exports 53.9; Ratio 79% - 2020: International ITO 46.3; Telecommunications, computer and information services exports 60.8; Ratio 76% - 2021: International ITO 55.0; Telecommunications, computer and information services exports 79.5; Ratio 69% - 2022: International ITO 56.1; Telecommunications, computer and information services exports 86.2; Ratio 65% - Note: International ITO accounts for 65 to 80 per cent of total exports of Telecommunications, computer and information services (EBOPS 2010 Item SI). #### Conclusions: lessons learned (MOFCOM) - Key contributing factors to success: - Statistical collection mandated by law: enterprises mandated to register business information under guidance of the National Bureau of Statistics of China. - Implementation of an online system: Online Monitoring System on Service Outsourcing established in 2009 and reviewed/upgraded every three years; user manual provided. - Regular training: nationwide trainings provided annually to local officials; local governments train enterprises; social-media contact group established. - Challenges: - Technical implementation and coverage: emerging business models are not always covered in time. - Scope limitation: currently international ITO covers only digitally delivered exports. - Further suggested improvements (MOFCOM plans): - Enhance and improve measurement of digitally delivered trade (specific steps are being considered in the source text). *Source: hmdtea - CHAPTER 6 — CASE STUDIES.* ### 1. Enhancing the Investigation System by: ### 1. Enhancing the Investigation System by: ### Enhancements to investigation forms and compilation - Requesting enterprises to provide information in the forms on the shares of digitally delivered services in the total services to facilitate the measurement of digital delivery of R&D, design, audio and video, as well as of creative services. - Adding digital intermediation platforms as a category to the forms to obtain the transaction values of digitally delivered trade through digital intermediation platforms from enterprises. - Compiling imports of services outsourcing. ### Diversifying data sources - MOFCOM aims to facilitate data-sharing with public data-holders such as China Customs, the State Taxation Administration and the Ministry of Industry and Information Technology. - MOFCOM will examine the feasibility of applying big data technology to fill in data gaps. ### Reinforcing international cooperation - MOFCOM aims to strengthen partnerships with international organizations and to exchange experiences and practices with other countries. ### Case study 3: Digital trade in Jamaica — Introduction - In 2021, Tax Administration Jamaica (TAJ) coordinated a stocktaking exercise to identify possible data sources for estimating digital trade in Jamaica, using official information and experimental data. - The exercise highlighted the feasibility of deriving estimates of digital trade by exploiting existing data sources without developing costly new survey instruments. - Most identified sources (living conditions survey, border surveys, payment card data, administrative sources) could be used to estimate digitally ordered trade. - Some progress has been made on measuring digitally delivered trade and trade via digital intermediation platforms; efforts are at an early stage. - The approaches described have not yet been implemented. - The stocktaking aimed to synchronize national data collection processes and leverage existing statistical datasets used for balance of payments and national accounts compilation. - The approach followed recommendations of the Handbook of Measuring Digital Trade (OECD, WTO, IMF, 2019). ### Data sources identified (summary) - Proposed national sources and the type of trade they can inform (Table 6.6): - Survey of Living Conditions (SLC) — Digitally ordered - Tourist Expenditure Survey (TES) — Digitally ordered - Tourist Satisfaction Survey (TSS) — Digitally ordered - Credit card data (Bank of Jamaica) — Digitally ordered/Digitally delivered - Administrative data: Tax Administration Jamaica (VAT tax forms) — Digitally ordered/Digitally delivered - Administrative data: Jamaican Customs Agency (customs data) — Digitally ordered ### Survey on Living Conditions (SLC) - Conducted once per year by STATIN via internet, regular mail and face-to-face interviews. - One question in Part L Section 6 asks whether the internet was used to purchase/order goods or services online (gives info on digitally ordered trade). - Additional questions on consumption of services (streaming movies, accessing educational and financial services) allow insights into digitally delivered trade. - STATIN is working on adapting the questionnaire to include information on the value of purchases made via the internet. - Limitations: - Does not provide monetary values; cannot identify digital trade transactions by value. - Goods and services are not separately identified in the digital ordering question. - No information to distinguish domestic vs international transactions. - Table 6.7: Digital trade dimensions in the SLC (coverage and notes) - Scope (Where?): Partial — Digital trade included but monetary value on digital flows is excluded. - Nature (How?): Partial — Digitally ordered partially covered; digitally delivered can be approximated. - Product (What?): No — Goods and services are not differentiated. - Actors (Who?): Partial — Households. - Flow (Sales/Purchases): Partial — Purchases covered. - Online channel: No — No distinction between different channels. - Geography (Domestic/Cross-border): No — No geographical breakdown for e-commerce into domestic and cross-border transactions. ### Tourist Expenditure Survey (TES) and Tourist Satisfaction Survey (TSS) - Conducted by Jamaica Tourist Board: TSS annually, TES bi-annually. - TSS provides information on booking channels for accommodation (insights into exports of digitally ordered services). - TES provides estimated expenditure of visitors on a monthly and seasonal basis, producing an annual estimate; breaks down short- and long-term stays and by residents, armed forces and non-resident Jamaicans. - TES covers inbound tourism flows (exports) only; resident tourism abroad could be assessed using additional sources such as credit card data. ### Credit card data (Bank of Jamaica) - BOJ collects information on credit card transactions of Jamaican residents; four international credit card companies (Visa, Mastercard, Japan Credit Bureau and American Express) required to submit data. - Credit card companies provide monthly data on expenditures by Jamaican residents cleared through foreign websites — a proxy for digitally ordered imports. - Current limitations (Table 6.8): - Scope (Where?): Partial — Data not disaggregated by product nor sector; partial disaggregation could be made available using payment gateway data. - Nature (How?): Partial — Expenditure items for both goods and services ordered via the internet; digitally delivered services can be identified based on the list of digitally deliverable services recommended in the Handbook. - Product (What?): Partial — Information on credit card transactions for goods and services is collected; classification by products/sectors is difficult. - Actors (Who?): No — Breakdown by residency not available; country of residence of issuing bank is reported rather than card-user residency. - Flow (Sales/Purchases): Partial — Only purchases. - Online channel: No — No distinction of type of online channel. - Industries: No — No industry breakdown. - Geography (Domestic/Cross-border): No — No geographical breakdown into domestic and cross-border transactions. - Future plans: - Use payment gateway data to better disaggregate credit card data by goods and services, obtain average transaction value and exact payment location. - Combine credit card data with payment gateway data — currently being investigated and not yet tested. ### Customs data (Jamaica Customs Agency, JCA) - JCA collects information on goods imported and exported and from couriers who deliver digitally ordered goods. - Based on current data, JCA estimates that over 90 per cent of the courier services are used for goods ordered from abroad. - Data compiled using UNCTAD ASYCUDA and business registers enable identification of enterprises registering courier services as core business; imports of such companies are summed and divided by total imports to estimate share of imports by courier services. - Shares of imports by courier services are included in balance of payments statistics and serve as a proxy for digitally ordered trade in goods. - Table 6.9: Extract of Jamaica balance of payments data enriched by customs data (Monthly 2020-21 data) — US$ millions - Goods and services (balance): Mar-20 -339.9; Jun-20 -691.6; Sep-20 -711.3; Dec-20 -825.5; Mar-21 -708.9; Jun-21 -579.9; Sep-21 -763.0; Dec-21 -976.7 - Credits: Mar-20 1465.2; Jun-20 482.2; Sep-20 630.6; Dec-20 764.9; Mar-21 850.5; Jun-21 1147.0; Sep-21 1180.8; Dec-21 1206.9 - Debits: Mar-20 1805.1; Jun-20 173.8; Sep-20 1341.9; Dec-20 1590.4; Mar-21 1559.4; Jun-21 1727.0; Sep-21 1943.8; Dec-21 2183.6 - Goods (balance): Mar-20 -853.3; Jun-20 -614.2; Sep-20 -660.0; Dec-20 -821.0; Mar-21 -647.2; Jun-21 -556.6; Sep-21 -577.9; Dec-21 -1043.9 - Exports (goods): Mar-20 358.2; Jun-20 264.7; Sep-20 306.4; Dec-20 321.2; Mar-21 360.3; Jun-21 422.9; Sep-21 371.5; Dec-21 286.0 - Imports (goods): Mar-20 1211.6; Jun-20 878.9; Sep-20 966.4; Dec-20 1142.2; Mar-21 1007.5; Jun-21 979.5; Sep-21 949.4; Dec-21 1329.9 - Services (balance): Mar-20 513.4; Jun-20 -77.4; Sep-20 -51.3; Dec-20 -4.5; Mar-21 -61.8; Jun-21 -23.3; Sep-21 -185.1; Dec-21 67.2 - Services credits: Mar-20 1107.0; Jun-20 217.5; Sep-20 324.2; Dec-20 443.8; Mar-21 490.1; Jun-21 724.2; Sep-21 809.3; Dec-21 920.9 - Services debits: Mar-20 593.6; Jun-20 294.9; Sep-20 375.4; Dec-20 448.3; Mar-21 551.9; Jun-21 747.5; Sep-21 994.4; Dec-21 853.7 - Total exports of goods: Mar-20 358.2; Jun-20 264.7; Sep-20 306.4; Dec-20 321.2; Mar-21 360.3; Jun-21 422.9; Sep-21 371.5; Dec-21 286.0 - Total imports of goods: Mar-20 1211.6; Jun-20 878.9; Sep-20 966.4; Dec-20 1142.2; Mar-21 1007.5; Jun-21 979.5; Sep-21 949.4; Dec-21 1329.9 - Total imports of couriers: Mar-20 122.0; Jun-20 115.7; Sep-20 106.6; Dec-20 84.0; Mar-21 156.6; Jun-21 118.5; Sep-21 118.5; Dec-21 45.7 - Courier imports as a % of total imports: Mar-20 10%; Jun-20 13%; Sep-20 11%; Dec-20 16%; Mar-21 16%; Jun-21 12%; Sep-21 12%; Dec-21 3% ### Administrative data on tax declarations (Tax Administration Jamaica) - TAJ collects VAT tax return information on services; revenue collected through tax payments on goods and services exported/imported by non-resident suppliers and resident importers are captured on returns (Tax Administration Jamaica, 2022). - Proposed uses: - Estimate proxies for digitally ordered and digitally delivered trade by matching values declared to lists of ICT-enabled goods (UNCTAD) and mapping enterprises to services categories (BPM6). - STATIN to derive a list of enterprises that typically sell digitally deliverable services and potentially impose an export threshold to identify delivered services. - Limitations and requirements: - Lack of product detail on tax declarations limits current use. - Tax declaration form should be adapted to indicate whether the product was digitally ordered or digitally delivered and provide clearer product descriptions. - Administrative data need mapping to services categories in BPM6 to derive proxies for share of services digitally delivered for exports. - An initial mapping is available in Annex G. ### Balance-of-payments Enterprise Survey - Quarterly Enterprise Survey is the main source to compile exports and imports of services by service item; sample includes 400 of the largest resident enterprises by asset size and gross profits. - Questionnaire administered via email or online portal; provides information on enterprise income and expenditure. - Does not explicitly distinguish digitally ordered vs digitally delivered services. - Possibility to derive proxies for digitally delivered services by considering EBOPS 2010 services items that are digitally deliverable (Table 6.10). - Business process outsourcing (BPO) services under "other business services" are particularly relevant and can be delivered digitally (Mode 1) or physically (Mode 4). - Combining survey data with the national statistical business register can provide first estimates on digitally delivered services for specific categories. - Table 6.10: Digital trade dimensions captured in the enterprise survey - Scope (Where?): Partial — Digitally delivered services. - Nature (How?): Partial — Digitally delivered services estimated by EBOPS 2010 items supplied by Mode 1 that are digitally deliverable. - Product (What?): Partial — Services according to EBOPS 2010. - Actors (Who?): Partial — No breakdown by client type; exports cover B2B, B2C and B2G; imports cover B2B and B2G. - Flow (Sales/Purchases): Total — Both exports and imports included. - Online channel: No — No information on channel (website, mobile app, digital intermediation platform). - Geography (Domestic/Cross-border): Total — Only cross-border services as Enterprise Survey covers resident vs non-resident transactions. - Industries: Total — All industries. ### Challenges and way forward - Jamaica intends to advance measurement of digital trade but faces methodological and implementation challenges. - Three key areas for future statistical improvements: 1. Reform the statistical law: Amend existing legislation on administrative data to improve measurement of digitally ordered trade (especially de minimis trade). 2. Improve measures of digitally delivered trade: Explore tax administrative data and BOJ efforts to measure cloud technology use and identify the source of imported services as components of digital trade. 3. Improve measurement of trade via digital intermediation platforms: BOJ to explore data from two payment gateways and international credit card processing to extract relevant information and enrich balance of payments statistics to derive a proxy measure for digital trade via digital intermediation platforms. *Handbook on Measuring Digital Trade — Chapter 6, Case study 3: Digital trade in Jamaica (extracted content).* ### 4.1 Introduction ### 4.1 Introduction ### Overview - INE performed a stock-taking exercise in 2022 to identify existing official statistics that could be leveraged to measure digital trade in Spain. - The approach builds on existing household and enterprise surveys, as well as border surveys, to reduce the burden on statistical offices and avoid duplication. - At the time of writing, the approach remains exploratory and has not yet been applied to statistical production processes at INE. - The approach could be applied to other members of the European Union (EU) because of methodologically homogeneous and comparable statistical sources across the EU. - Sections 4.2 and 4.3 demonstrate how existing sources can be used to gather insights related to digitally ordered and digitally delivered trade. Section 4.4 summarizes strengths and weaknesses. ### Data sources assessed (high-level) - Enterprise surveys assessed include: - Survey on Information and Communications Technology (ICT) Usage and E‑commerce in Enterprises (ICT-E) - Statistics on Products in the Trade Sector (SPTS): Retail module - International Trade in Services Survey (ITSS) — noted as related to digitally delivered trade - Household surveys assessed include: - Survey on Equipment and Use of Information and Communication Technologies in Households (ICT-H) - Household Budget Survey (HBS) - Residents Travel Survey (RTS) - Border surveys assessed include: - Tourist Expenditure Survey (EGATUR) - Tourist Movement on Borders Survey (FRONTUR) ### Key survey characteristics and sample sizes (as reported) - ICT-E: - For companies with 10 or more employees: around 15,000 companies included. - For companies with less than 10 employees: around 11,000 companies included. - Collected via the internet and regular mail. - SPTS (retail module): - Sample size includes approximately 60,500 companies. - Collected with an online questionnaire. - ICT-H: - Sample includes 2,500 census sections, corresponding to around 25,000 dwellings. - Annual household survey carried out since 2002. - HBS: - Conducted every five years. - Sample size is approximately 24,000 households per year. - RTS: - Provides monthly, quarterly and annual estimates of trips by Spanish residents (used with credit card information for travel debits). - FRONTUR and EGATUR: - Integrated to provide statistics on tourist movements and tourist expenditure; share the same questionnaire. ### ICT-E: scope, coverage and key limitations - Coverage and temporal scope: - ICT use variables refer to the first quarter of the year of the survey; general company information, electronic commerce and ICT training refer to the previous year. - E‑commerce coverage: - Captures digitally ordered trade (e-commerce) for goods and services per the Handbook definition. - Geographic breakdown: domestic (Spain), intra-EU and rest of the world — allows e-commerce imports and exports to be identified. - Sales channel: imports broken down by online store, web application or electronic data interchange (EDI); exports disaggregated into web exports (website, application or marketplace) and EDI. - Key limitation: - The survey does not distinguish between goods and services. - The questionnaire structure inhibits cross-panel analysis across certain variables (for example, combining e-commerce exports, actors per transactions, digital intermediation platforms (DIPs) and geography is not possible because information is not collected for all dimensions). ### ICT-E: selected e‑commerce figures retrieved (Survey on ICT usage and e‑commerce in enterprises (2019‑20)) - Total enterprises: - E-commerce sales (1000 of EUR): 307,424,523 - % e-commerce sales on total sales: 18.34 - % e-commerce sales within Spain on total e-commerce sales: 81.27 - % e-commerce sales to the rest of the EU on total e-commerce sales: 13.94 - % e-commerce sales to the rest of the world on total e-commerce sales: 4.78 - Total industry (NACE 10-39): - E-commerce sales (1000 of EUR): 158,067,721 - % e-commerce sales on total sales: 24.52 - % e-commerce sales within Spain on total e-commerce sales: 80.62 - % e-commerce sales to the rest of the EU on total e-commerce sales: 15.02 - % e-commerce sales to the rest of the world on total e-commerce sales: 4.36 - Selected subcomponents (E‑commerce sales (1000 of EUR) and % e‑commerce sales on total sales): - NACE 10-18: 32,785,664; 21.17 - NACE 19-23: 56,160,480; 33.92 - NACE 24-25: 5,451,762; 7.98 - NACE 26-33: 43,342,464; 29.27 - NACE 35-39: 21,327,351; 19.29 - Total services (NACE 45-82, excluding NACE 56, 66, 75 and 64-66): - E-commerce sales (1000 of EUR): 148,047,863 - % e-commerce sales on total sales: 17.37 - ICT sector (261-264, 268, 465, 582, 61, 6201, 6202, 6203, 6209, 631, 951): - E-commerce sales (1000 of EUR): 14,705,099 - % e-commerce sales on total sales: 17.41 ### SPTS (Retail module): scope, coverage and limitations - SPTS retail module: - Yearly structural survey covering wholesale and retail trade (section G of NACE Rev.2, “trade” sector). - Includes turnover breakdown by product and by sales system (which includes e-commerce). - Total e-commerce turnover is broken down by broad product category and by type of customer (B2B vs B2C). - Key limitations: - Cross-border dimension is not directly covered. - It is not possible to extract cross-collected information on e-commerce sales by actor, sale channel, geographic breakdown or product simultaneously. - E‑commerce sales estimated from the retail module for NACE G47 should be checked against ICT-E for G47 to avoid double counting. ### Household surveys: ICT-H and HBS — scope and limitations - ICT-H: - Module VII provides information on e-commerce purchases (imports) only, covering households and resident natural persons. - Distinguishes goods and services with a product breakdown compatible with CPA. - Geographic breakdown: domestic and cross-border (rest of EU and rest of world) — allows cross-border e-commerce imports to be distinguished from domestic purchases. - Actor breakdown: B2C and C2C provided for some services. - Key limitations: almost all e-commerce data are qualitative; only limited quantitative information (question on total value of e-commerce in the last three months). No breakdown by e-commerce channel (website, mobile app or DIP). - HBS: - Provides information on e-commerce purchases for many consumption products according to ECOICOP. - Goods and services can be distinguished and ECOICOP items can be converted to CPA or EBOPS 2010. - Key limitations: lack of geographical breakdown prevents distinguishing cross-border from domestic e-commerce; no breakdown by channel or actors. ### Travel and border surveys: RTS, FRONTUR and EGATUR — scope and limitations - RTS: - Provides monthly, quarterly, annual estimates of resident trips and characteristics (destination, duration, purpose, expenditure, etc.). - Collects reservation channel used to book transport, accommodation, tour packages and other digitally ordered travel-related goods and services. - As a household survey covering residents, it distinguishes domestic and cross-border tourism services for residents but only covers imports (partial coverage of digitally ordered trade). - FRONTUR and EGATUR: - Integrated to measure non-resident visitors arriving in Spain and their expenditure; share the same questionnaire. - Offer insights on the booking channel for accommodation and travel (digitally ordered trade) for non-residents; cover travel credits in balance of payments estimation. ### Summary of digital trade dimension coverage across key surveys (selective) - ICT-E (enterprises): - Scope (Where?): Partial — digital trade included in conventional trade/business statistics; non-monetary digital flows excluded. - Nature (How?): Partial — digitally ordered covered; goods and services not differentiated. - Product (What?): No — goods and services not differentiated. - Actors (Who?): Partial — businesses; B2B/B2C/B2G breakdown for sales. - Flow (Sales/Purchases): Partial — both included but more dimensions for sales than for purchases. - Online channel: Total — web/app, DIP and EDI distinguished. - Geography: Total — domestic, rest of EU, rest of world. - Industries: Partial — covers C, D, E, F, G, H, I, J, L, M(69-74), N, S(95.1); financial corporations excluded. - SPTS (retail module): - Product: Partial — only goods (retailers). - Channel: Partial — distinguishes e-commerce vs offline but not web/app vs DIP vs EDI. - Geography: No — geographic breakdown not directly observable but could be estimated. - ICT-H and HBS (households): - Generally provide product-level information and can identify digitally deliverable services (for some services), but commonly lack channel and actor breakdowns and (for HBS) geographic origin of e-commerce purchases. ### Strengths and weaknesses (as identified) - Strengths: - Existing surveys already collect many relevant dimensions (e.g., channel, geographic breakdown in some enterprise and household surveys, actor types for some modules). - Methodologically harmonized EU surveys (e.g., ICT-E, ICT-H) facilitate comparability across EU members. - Integration of border and travel surveys supports balance of payments estimates for travel credits and debits. - Weaknesses: - Important gaps remain: separation of goods vs services in enterprise ICT-E; limited quantitative monetary detail in household ICT-H; limited simultaneous cross-dimension linkage (e.g., channel × actor × geography × platform). - Potential for double counting (e.g., e-commerce sales for G47 between ICT-E and SPTS) requires reconciliation. - Non-monetary digital flows are excluded across conventional sources. *Source: hmdtea - 4.1 Introduction (INE stock-taking exercise and assessment of Spanish data sources for measuring digital trade).* ### CHAPTER 6 — CASE STUDIES ### CHAPTER 6 — CASE STUDIES ### 4.3 Data sources related to digitally delivered trade - International Trade in Services Survey (quarterly business survey) - Sample includes 9,500 resident enterprises. - Captures exports and imports of services by service item, partner country, modes of supply, main economic activity and number of employees. - Questionnaire is directly addressed to companies via mail, fax or e-mail or by means of the IRIA software application (Instituto Nacional de Estadística, 2015). - Collects information on digitally deliverable services supplied via Mode 1. - For all services which are digitally deliverable, Mode 1 supply constitutes a reasonable proxy for digital delivery. - Economic activity variable and industry classifications (NACE Rev.2 and ISIC Rev.4) allow refined analysis to identify active traders in ICT sectors and in high- and medium-high-technology industries that produce knowledge-intensive services according to NACE Rev.2. - The questionnaire does not capture digital ordering. - Coverage notes from related surveys (summarized) - FRONTUR/EGATUR (inbound tourism) focuses on tourism expenditure; limited number of tourism products (mostly services) considered. - RTS/household surveys include digitally ordered tourism expenditure components (accommodation, transport, package tours) but non-monetary digital flows are excluded. - Many sources can provide product-level and flow (exports/imports) information; ICT-E only provides aggregate products for some flows. ### Key statistics: Share of exports/imports of services supplied by Mode 1 (International Trade in Services Survey 2014‑19, in percentages) - Total: Exports 85.3, Imports 94.5 - 1. Manufacturing services on physical inputs owned by others: Exports 5.1, Imports 14.4 - 2. Maintenance and repairs: Exports 1.8, Imports 6.3 - 3. Transport: Exports 86.0, Imports 96.6 - 4. Construction: Exports 35.0, Imports 36.8 - 5. Insurance services: Exports 95.7, Imports 95.5 - 6. Financial services: Exports 99.5, Imports 99.8 - 7. Charges for the use of intellectual property: Exports 99.6, Imports 98.5 - 8. Telecommunications, computer and information services: Exports 98.1, Imports 96.9 - 9. Other business services: Exports 91.2, Imports 97.9 - 9.1 Research and development: Exports 99.4, Imports 99.2 - 9.2 Professional and management consultancy: Exports 94.0, Imports 98.4 - 9.3 Technical, trade-related and other business services: Exports 88.9, Imports 97.5 - 10. Personal, cultural and recreational services: Exports 92.3, Imports 74.9 - 11. Government goods and services n.i.e.: Exports 90.8, Imports 28.9 ### 4.4 Conclusions (Spain case study) - Overall assessment - Spanish statistical system provides a range of data sources usable to generate meaningful estimates of digital trade. - Digitally ordered trade can be identified using several different data sources. - Digitally delivered trade estimates can be largely derived from the existing international trade in services survey. - Strengths - Multiple sources allow measurement of total e-commerce (domestic and cross-border). - Some sources provide explicit cross-border dimensions; many provide product (goods/services) and flow (exports/imports) breakdowns. - Mix of surveys covers actors across corporations, households and NPISHs. - Limitations and gaps - ICT-E only provides aggregate products; household and border surveys mainly record the import side. - Some surveys do not include government transactions except via EBOPS 2010 item “Government goods and services n.i.e.” - Surveys were not originally designed to measure digital trade; limitations include: - Imbalance in coverage of digital trade across surveys (enterprises vs households vs border surveys). - Difficulty capturing both delivery and ordering dimensions for services. - Lack of granularity and geographical breakdowns (domestic vs cross-border) in some surveys. - Insufficient information on transactions facilitated by DIPs; ICT usage/e-commerce surveys include DIPs only partially. - Not possible to derive estimates of intermediation services provided by DIPs without additional targeted questions. - Data validation and reconciliation recommendations - Validate and reconcile sources to ensure robustness and avoid duplication. - Compare e-commerce exports/imports to natural persons (B2C) from household and border surveys with alternative data-source estimates. - Check enterprise e-commerce purchases (imports) against sales (exports) for supply-demand coherence. - Decide which data source is best suited to provide reliable estimates before final compilation. ### Case study 5: Measuring digitally ordered merchandise trade in Türkiye #### 5.1 Introduction - Context - Official international trade in goods statistics in Türkiye are produced in cooperation between TurkStat and the Ministry of Trade; cover general and special trade systems and are collected by the Ministry of Trade using administrative records. - Official statistics on trade in goods do not currently specify the share of digitally ordered trade. - Pilot study - TurkStat and the Ministry of Trade carried out a pilot study to produce estimates of digitally ordered trade (as defined in this Handbook), focusing on merchandise trade. - Methodology uses administrative records: customs data, electronic customs declarations and postal data, with quality and robustness checks at country/product levels. - At time of writing, approach still being tested; official estimates at country, sector and product-level are planned for release by the end of 2023. #### 5.2 Data sources (Türkiye) - Methodological guidance - TurkStat examined available sources following UNCTAD Manual for the Production of Statistics on the Digital Economy (UNCTAD, 2020) and previous Handbook version (OECD, WTO, IMF, 2019). - Three administrative sources identified as suitable: customs declarations (detailed declarations), electronic trade customs declarations (ETCDs), and postal services data. - 5.2.1 Customs declarations (detailed declarations) - Customs records record all kinds of cross-border trade in goods; Ministry of Trade collects these records. - To identify digitally ordered transactions, the question “Is this import/export electronic trade (e-trade)?” was added to the declaration form in 2019 and revised in December 2021. - Within the pilot, responses were analysed and businesses completing declarations were interviewed for feedback. - Ministry of Trade and TurkStat prepared a guidance note and a training video in Turkish to help customs counsellors in companies completing declarations (video can be accessed at https://tagm.ticaret.gov.tr/duyurular/sinir-otesi-elektronik-ticaret-e-ticaret-rehberi-ve-tanitim-videosu-yayinlandi). - Since the beginning of 2022, data on digitally ordered trade identified via customs declarations were directly provided to TurkStat. - Data quality measures and processing - Data mining methods, initial clustering analysis and outlier detection approaches used to ensure data quality. - Variables such as customs regimes and products used to control accuracy of declarations flagged as digital orders. - Manual checks at product and firm level used to verify submitted declarations. - Questionnaire design includes control questions to designate whether the product is digitally ordered; companies were contacted if answers were missing or incorrect. - Product-specific patterns in inaccurate declarations were observed (e.g., aerospace products or live animals rarely ordered digitally and required manual verification). - Observation: custom declarations of B2C products with high value are usually not traded digitally; checks help monitor correct completion rates and optimize collection/verification processes. - 5.2.2 Electronic Trade Customs Declarations (ETCDs) - ETCDs have been used since 2012; useful to identify digital trade transactions at micro level. - ETCDs issued electronically by authorized express airline cargo companies; transactions eligible for electronic declarations have upper limits: - 15,000 euros/300 kilogrammes for exports - 150 euros/30 kilogrammes for imports - Since the beginning of 2021, ETCD data are available and include: country of origin (for imports), destination country (for exports), value, quantity and product (up to HS six-digit codes) breakdowns. - New legal and IT infrastructure allows Ministry of Trade to provide ETCD data directly to TurkStat. - Outlier detection methods added in the data processing system to detect erroneous records. - 5.2.3 Data from postal services - Electronic trade customs declaration processes could be improved to capture digitally ordered trade from postal services. - General Directorate of Post and Telegraph Organization (PTT) completes ETCD for exports. - For imports, declarations do not follow ETCD standards yet, as they are completed by authorities of partner countries. - PTT is exploring harmonization of declaration processes coming from different countries. - Data compilation from postal channels was still pending at time of writing. *Source: CHAPTER 6 — CASE STUDIES, Handbook on Measuring Digital Trade* ### 5.3 Preliminary results ### hmdtea - 5.3 Preliminary results ### Preliminary results (ETCDs) - Context: - The approach is currently being tested; this case study shows preliminary aggregate results for a small sample of digitally ordered statistics obtained via ETCDs, the second channel used by the Ministry of Trade to measure digitally ordered trade. - TurkStat plans to publish complete statistics on digitally ordered trade by the end of 2023. - Key statistics (according to preliminary results obtained from ETCDs): - The monthly average value of exports is US$ 110 million. - The monthly average number of export transactions is 1.4 million. - The monthly average value of imports is US$ 16 million. - The monthly average number of import transactions is 0.3 million. - The share of total ETCD exports (by value) in total exports is 0.58 per cent. - The share of total number of ETCD export transactions in total exports transactions is 41 per cent. - The share of total ETCD imports (by value) in total imports is 0.07 per cent. - The share of total number of ETCD imports transactions in total imports transactions is 21.3 per cent. - The average transaction value for exports is US$ 78. - The average transaction value for imports is US$ 55. ### Conclusions, identified challenges, and quality measures - Achievements: - TurkStat has been successful in leveraging customs and postal data to separately identify digitally ordered transactions in its merchandise trade statistics. - Building infrastructure and workflow to compile digital trade statistics provided valuable lessons useful to other compilers. - Limitations and challenges identified: - Transactions performed by enterprises via EDI cannot be detected; a new question currently being tested in the customs declaration system may help identify digitally ordered trade via EDI in the future. - Transactions conducted via DIPs cannot be identified. - B2B, B2C and C2C transactions cannot be distinguished separately. - In some cases product codes are not declared at a sufficiently detailed level. - Data quality and capacity-building measures already undertaken: - Development of a clear understanding of the definition of cross-border digital trade statistics and determination of scope using practical scenarios and examples is key. - Ensuring that people who fill out customs declarations have sufficient knowledge about digital trade (specifically “digital ordering”) is necessary. - The Ministry of Trade and TurkStat have provided extensive material to raise awareness and educate custom counsellors, including a training video and a specific guide. - Extensive quality assurance processes are indispensable to verify the submitted customs declarations. ### Planned future work and next steps - TurkStat plans the following future steps: 1. To increase the coverage of all data sources. 2. To implement measures to ensure that customs declarations are completed correctly. The Ministry of Trade plans to conduct study visits to customs officials that file custom declarations registering a high value of digitally ordered goods to harmonize the data collection process. Additional training is also planned for the officials responsible for completing the declarations. Moreover, measures are being considered to penalize companies which consistently complete declarations wrongly. 3. To analyse the data after its compilation across all three sources (i.e., customs records, ETCDs and data from postal services) and obtain initial results. Currently, the objective is to generate a coherent time series and deploy quality assurance checks. 4. To carry out the necessary research to enable further breakdowns of B2B, B2C and C2C transactions. 5. To carry out studies of how to include digital intermediation platforms in the compilation of trade data. *Source: hmdtea - 5.3 Preliminary results — https://www.imf.org/-/media/files/publications/books/2023/english/hmdtea.pdf* ### Annex B: Accounting ### Annex B: Accounting for digital trade transactions ### Understanding the annex table - This annex has been included to help compilers use the “Reporting template for digital trade” (Table 2.1 in Chapter 2), reproduced below. *Source: hmdtea - Annex B: Accounting* ### Annex Table B.1, included hereafter, enumerates the ### Annex Table B.1 ### Purpose and scope - Annex Table B.1 enumerates: what is being transacted (i.e., goods or services); the nature of the transaction (i.e., digitally ordered and/or digitally delivered); and who the parties involved in the transaction are (i.e., business-to-business (B2B), business-to-consumer (B2C), etc.). - The annex table demonstrates how digital trade transactions should be recorded in the “Reporting template for digital trade” (Table 2.1). - Examples are provided (numbered 1–35) to help interpret and apply the recording of transactions in the reporting template. References to the template item(s) according to which any given transaction should be recorded are provided in parentheses. ### Reporting template overview (Table 2.1) - Table 2.1 (CHAPTER 2): REPORTING TEMPLATE FOR DIGITAL TRADE - Items and formulas: - Item 1: Total digital trade — 2+3 minus 4 - Item 2: Digitally ordered trade — 2.1+2.2 - Item 2.1: Goods - Item 2.1.a: of which: via DIPs - Item 2.2: Services - Item 2.2.a: of which: via DIPs - Item 3: Digitally delivered trade - Item 3.a: of which: via DIPs - Item 4: Digitally ordered and digitally delivered trade - Item 4.a: of which: digital intermediation services - Addendum items: - A.1 Digital trade in services — 2.2+3 minus 4 - A.2 Digitally deliverable services — >3 ### Treatment of digital intermediation services - Digital intermediation services are, by definition, both digitally ordered and digitally delivered, and should be recorded in: - Item 2.2: “Digitally ordered trade; Services” - Item 3: “Digitally delivered trade” - Item 4: “Services digitally ordered and digitally delivered” - Item 4.a: “Services digitally ordered and digitally delivered; Of which digital intermediation services” - Digital intermediation services are provided by the digital intermediation platform (DIP) to both the buyer and the seller. - Fee payer treatment: - The fee for the digital intermediation service may be paid by the seller, the buyer, or both. - If it is not known who paid the fee, the Handbook recommends recording the fee as having been paid by the seller. This can affect which countries need to record the digital intermediation service when completing the reporting template. - Additional guidance and worked example references: - Additional guidance on recording the digital intermediation service fee is included in Chapter 5, Table 5.3. - A fully worked-up example, including statistics, can be viewed in Table 5.2, within Box 5.1. ### Recording guidance and example patterns (summary of Examples 1–35) - General recording principle: - Some transactions must be recorded in more than one line of the reporting template. For example, when a transaction is both digitally ordered (item 2) and digitally delivered (item 3), it must be subtracted (item 4) to avoid double-counting when calculating total digital trade (item 1). - Typical example patterns (referenced by example number): - Examples where digitally ordered only (YNN) for goods or services → record in 2.1 or 2.2 (Examples 1–4). - Transactions where goods or services are ordered via DIP but supplier is resident or non-resident and DIP location varies (YYN, YYY, YNY, etc.) → combinations of 2.1, 2.1.a, 2.2, 2.2.a, 3, 3.a, 4, 4.a depending on whether digital intermediation services are considered international (Examples 5–21, 24–31). - Transactions both digitally ordered and digitally delivered (YNY) → record in 2.2, 3, and 4 (Examples 22–23). - Offline order with digital delivery (NNY) → record in 3 only (Examples 34–35). - Consumer-to-consumer transactions via non-resident DIPs (YYNC, YYYC, YYY) → may generate digitally ordered exports/imports and digital intermediation services imports/exports depending on DIP country and fee payer (Examples 17, 21, 32–33). - Specific illustrative behaviors captured across examples: - When buyer and seller are in the same country and DIP is non-resident, often only digital intermediation services are recorded as international trade (Examples 5, 10, 14, 24, 28, 33). - When DIP is resident and supplier non-resident, digitally ordered goods/services are recorded as imports/exports and digital intermediation services may be recorded as imports/exports depending on payer (Examples 6, 11, 15, 25). - When DIP is located in a third country (Country C), digital intermediation services are recorded as imports from Country C and potentially as exports to Country A and/or Country B (Examples 8, 9, 13, 21, 27). ### Classification, compilation flexibility, and product breakdown - Note: Transactions should be broken down by relevant product groupings (Extended Balance of Payments Services Classification (EBOPS) 2010 for services and, for example, the Harmonized Commodity Description and Coding System (HS) or Central Product Classification (CPC) for goods). - Compilers have flexibility to populate the template according to sources and details available. It is not essential to have separate measures of transactions involving digital intermediation platforms in order to record overall “digitally ordered trade” or “digitally deliverable trade”, or to calculate “total digital trade”. *Source: IMF, OECD, UNCTAD and WTO.* ### Annex C ### Annex C ### SDMX BOP DSD to EBOPS 2010 mapping — insurance, financial, IP, telecom, computer, information, R&D, professional, and other services - Insurance and pension servicesSF6 - Direct insuranceSF16.1 - Life insuranceSF116.1.1 - Freight insuranceSF126.1.2 - Other direct insuranceSF136.1.3 - ReinsuranceSF26.2 - Auxiliary insurance servicesSF36.3 - Pension and standardized guarantee servicesSF46.4 - Pension servicesSF416.4.1 - Standardized guarantee servicesSF426.4.2 - Financial servicesSG7 - Explicitly charged and other financial servicesSG17.1 - Financial intermediation services indirectly measured (FISIM)SG27.2 - Charges for the use of intellectual property n.i.e.SH8 - Franchises and trademarks licensing feesSH18.1 - Licences for the use of outcomes of research and developmentSH28.2 - Licences to reproduce and/or distribute computer softwareSH38.3 - Licences to reproduce and/or distribute audio-visual and related productsSH48.4 - Licences to reproduce and/or distribute audio-visual productsSH418.4.1 - Licences to reproduce and/or distribute other productsSH428.4.2 - Telecommunications, computer, and information servicesSI9 - Telecommunications servicesSI19.1 - Computer servicesSI29.2 - Computer softwareSI219.2.1 - Of which: Software originalsSI21z9.2.1.a - Other computer servicesSI229.2.2 - Information servicesSI39.3 - News agency servicesSI319.3.1 - Other information servicesSI329.3.2 ### Research and development; professional, technical, business, and personal services - Research and development servicesSJ110.1 - Work undertaken on a systematic basis to increase the stock ofSJ1110.1.1 - Provision of customized and non-customized research and development servicesSJ11110.1.1.1 - Sale of proprietary rights arising from research and developmentSJ11210.1.1.2 - PatentsSJ112110.1.1.2.1 - Copyrights arising from research and developmentSJ112210.1.1.2.2 - Industrial processes and designsSJ112310.1.1.2.3 - Other sales of proprietary rights arising from research and developmentSJ112410.1.1.2.4 - Other research and development servicesSJ1210.1.2 - Professional and management consulting servicesSJ210.2 - Legal, accounting, management consulting and public relations servicesSJ2110.2.1 - Legal servicesSJ21110.2.1.1 - Accounting, auditing, bookkeeping and tax consulting servicesSJ21210.2.1.2 - Business and management consulting and public relations servicesSJ21310.2.1.3 - Advertising, market research and public opinion polling servicesSJ2210.2.2 - Of which: Convention, trade fair and exhibition organization servicesSJ22z10.2.2.1 - Architectural, engineering, scientific and other technical servicesSJ3110.3.1 - Architectural servicesSJ31110.3.1.1 - Engineering servicesSJ31210.3.1.2 - Scientific and other technical servicesSJ31310.3.1.3 - Trade-related servicesSJ3410.3.4 - Other business services n.i.e.SJ3510.3.5 - Of which: Employment services, i.e., search, placement and supply of personnel services SJ35z10.3.5.1 - Audio-visual and related servicesSK111.1 - Audio-visual servicesSK1111.1.1 - Of which: Audio-visual originalsSK11z11.1.1.a - Artistic-related servicesSK1211.1.2 - Health servicesSK2111.2.1 - Education servicesSK2211.2.2 - Heritage and recreational servicesSK2311.2.3 ### Table C.1 context — digitally deliverable services and travel presentation - Table C.1: EXPANDED TABLE ON DIGITALLY DELIVERABLE SERVICES IN EBOPS 2010 - Digitally deliverable services consumed abroad (Mode 2) are recoded under Travel (SD) - Travel: by purpose of travel - BusinessSDA4.1 - Acquisition of goods and services by border, seasonal and other short-term workers SDA14.1.1 - Other (Business travel)SDA24.1.2 - PersonalSDB4.2 - Health-relatedSDB14.2.1 - Education-relatedSDB24.2.2 - Other (Personal travel)SDB34.2.3 - Travel - alternative presentation: by product - Other servicesSD54a.5 - Of which: Health servicesSD5z4a.5.1 - Of which: Education servicesSD5y4a.5.2 *Source: Handbook on Measuring Digital Trade — Annex C.* ### 4. Have you purchased any of the above-mentioned ### hmdtea - 4. Have you purchased any of the above-mentioned ### Survey questions on online purchases and hiring (private use) - Question 4: Have you purchased any of the above-mentioned products from an individual via a website or an application (e.g., via eBay or Facebook Marketplace)? - Possible responses: Yes, in the last three months; Yes, more than three months and less than one year ago; No - Question 5 (only for products ordered or purchased during the last three months): From whom did you buy them? (Include online purchases from companies and individuals) - Possible responses: Yes/No - a) National sellers (including national websites of multinational companies, e.g., Amazon Spain) - b) Vendors from other EU countries - c) Vendors from the rest of the world - d) The vendor’s country of origin is unknown - Question 6: Have you purchased or subscribed to any of the following products via a website or application for private use in the last three months? Free applications are excluded. - Possible responses: Yes/No - a) Streaming or downloaded music - b) Streaming or downloaded films or television series - c) Online games, including games downloaded for smartphones, computers, tablets or consoles - d) Downloads of software, including updates - e) Applications related to health or to physical fitness - f) Other applications (e.g., related to language learning or travel) - Question 7: Have you purchased or subscribed to any of the following products via a website or application for private use in the last three months? - Possible responses: - a) Tickets for sports events - b) Tickets for cultural or other events (movies, concert, fairs, etc.) - c) Internet subscriptions or mobile phone connections - d) Subscriptions for electricity, water or gas supplies - Question 8: For private purposes, have you hired (contact and agreement), via a website or application, any home service (e.g., cleaning, childcare, repair work, gardening)? - Possible responses: Yes, in the last three months; Yes, more than three months and less than one year ago; No - a) With a service provider - b) With an individual (e.g., via Facebook Marketplace) - Question 9: Have you used a website or application to hire a transport service (a car, for example) for private purposes? - Possible responses: Yes, in the last three months; Yes, more than three months and less than one year ago; No - a) With a transport company, bus company, flight company, taxi service or driver (e.g., Cabify, Uber, Free Now) - b) With an individual (e.g., Blablacar, Amovens) - Question 10: Have you used a website or application to book accommodation (room, apartment, house, etc.) for private purposes? - Possible responses: Yes, in the last three months; Yes, more than three months and less than one year ago; No - a) With a company, such as a hotel or travel agency - b) With an individual (e.g., Airbnb, HomeAway) - Question 11: Have you hired any other service not already mentioned, for private purposes (excluding financial and insurance services) via a website or application? - Possible responses: Yes, in the last three months; Yes, more than three months and less than one year ago; No ### Table H.1 — Results of the ICT-H survey: Internet purchases by demographic characteristics and range of value (2020) - Table heading: Total persons (16‑74 years) (number of people and percentage) — People who made purchases via the internet in the last three months (number of people and percentage) - Entries as presented in source: - Total persons35,238,28818,965,977 - Less than EUR 509.6%17.9% - EUR 50 to less than EUR 10011.4%21.2% - EUR 100 to less than EUR 30015.0%27.9% - EUR 300 to less than EUR 5007.2%13.4% - EUR 700 to less than EUR 10002.9%5.3% - More than 1000 EUR1.5%2.8% - Unknown/Non-response2.5%4.7% - No purchases made via the internet in the last three months3.7%6.9% - Source notation in table: Source: INE. ### Purchase frequency, total value categories, and financial activities (questions 12–14) - Question 12: How often have you purchased or ordered goods or services via the internet for private purposes in the last three months? - Possible responses: - a) Once or twice - b) Three to five times - c) Six to ten times - d) More than ten times - Question 13: Excluding the purchase of shares and other financial assets, what was the total value of the goods and services you have purchased or ordered via the internet for private purposes in the last three months? - Possible responses: - a) Less than €50 - b) From €50 to less than €100 - c) From €100 to less than €300 - d) From €300 to less than €500 - e) From €500 to less than €700 - f) From €700 to less than €1,000 - g) More than €1,000 - h) Unknown/No response - Question 14: Have you performed any of the following financial activities via the internet for private purposes in the last three months (excluding transactions via e-mail)? - Possible responses: Yes/No - a) Purchase or sale of shares, bonds, funds or other financial investment products - b) Underwriting of insurance policies or renewal of existing ones, including those offered as a package together with another service (e.g., travel insurance offered together with a flight ticket) - c) Drawing up a loan or mortgage contract or obtaining credit from a bank or other financial supplier ### Endnotes and bibliographic context (as included in source) - Endnotes summary: - Endnote 1: Annex based on the OECD Handbook on Compiling Digital Supply and Use Tables (OECD, 2023). - Endnote 2: ICT goods and digital services correspond to the CPC Rev.2.1 list of ICT products (United Nations, 2015). - Endnote 3: The Ministerial Declaration can be accessed at the stated URL. - Endnote 4: Documentation on the SNA revision process is available at the stated URL. - Endnote 5: For a detailed definition of each industry, see OECD Handbook on Digital Supply and Use Tables (2023). - Endnote 6: For instance, Canada, the Netherlands and Ireland. - Endnote 7: See the stated Guidance Note DZ.5 URL. - Endnote 8: Forms - TAJ Portal (jamaicatax.gov.jm): stated URL. - Bibliography: The chapter/annex includes an extensive bibliography of methodological and empirical references relevant to measuring digital trade and digital economy statistics (authors and titles listed in the source). *Handbook on Measuring Digital Trade — Annexes (excerpt provided)* ### BIBLIOGRAPHY ### BIBLIOGRAPHY ### OECD and related OECD/G20 outputs - Organisation for Economic Co-operation and Development (OECD) (2011), OECD Guide to Measuring the Information Society 2011, Paris: OECD Publishing. Retrieved from https://dx.doi.org/10.1787/9789264113541-en. - Organisation for Economic Co-operation and Development (OECD) (2015a), “The OECD Model Survey on ICT Usage by Businesses (2nd revision)”, Paris: OECD. Retrieved from https://www.oecd.org/sti/ieconomy/ICT-Model-Survey-Usage-Businesses.pdf. - Organisation for Economic Co-operation and Development (OECD) (2015b), The OECD Model Survey on ICT Access and Usage by Households and Individuals. Retrieved from https://www.oecd.org/sti/ieconomy/ICT-Model-Survey-Access-Usage-Households-Individuals.pdf. - Organisation for Economic Co-operation and Development (OECD) (2016), “Results of the 2016 WPTGS stocktaking questionnaire”, Paris: OECD. Retrieved from https://one.oecd.org/document/STD/CSSP/WPTGS(2016)7/en/pdf. - Organisation for Economic Co-operation and Development (OECD) (2018a), Measuring online platforms and cloud computing in National Accounts. Workshop: Working Party on Measurement and Analysis of the Digital Economy. Retrieved from https://www.oecd.org/going-digital/mdt-roadm-ap-cloud-computing.pdf. - Organisation for Economic Co-operation and Development (OECD) (2018b), Tax Challenges Arising from Digitalisation – Interim Report 2018: Inclusive Framework on BEPS, OECD/G20 Base Erosion and Profit Shifting Project, Paris: OECD Publishing. Retrieved from https://dx.doi.org/10.1787/9789264293083-en. - Organisation for Economic Co-operation and Development (OECD) (2018c), “Working Party on International Trade in Goods and Trade in Services Statistics – Result of the 2018 WPTGS Stocktaking Questionnaire”, Paris: OECD. Retrieved at https://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=STD/CSSP/WPTGS(2018)3&docLanguage=En. - Organisation for Economic Co-operation and Development (OECD) (2019a). An Introduction to Online Platforms and Their Role in the Digital Transformation, Paris: OECD Publishing. Retrieved from https://doi.org/10.1787/53e5f593-en. - Organisation for Economic Co-operation and Development (OECD) (2019b), Going Digital: Shaping Policies, Improving Lives, Paris: OECD Publishing. Retrieved from https://doi.org/10.1787/9789264312012-en. - Organisation for Economic Co-operation and Development (OECD) (2019c), “WCO cross-border e-commerce framework of standards (2018)”, in The Role of Digital Platforms in the Collection of VAT/GST on Online Sales, Paris: OECD Publishing. Retrieved from https://doi.org/10.1787/8c36d673-en. - Organisation for Economic Co-operation and Development (OECD) (2019d), “BBVA big data on online credit card transactions: The patterns of domestic and cross-border e-commerce”, OECD Digital Economy Papers, No. 278, Paris: OECD. Retrieved from https://doi.org/10.1787/8c408f92-en. - Organisation for Economic Co-operation and Development (OECD) (2020), A roadmap toward a common framework for measuring the Digital Economy: Report for the G20 Digital Economy Task Force, Paris: OECD Publishing. Retrieved from http://www.oecd.org/sti/roadmap-toward-a-common-frame-work-for-measuring-the-digital-economy.pdf. - Organisation for Economic Co-operation and Development (OECD) (2021), “Summary record of the Working Party on International Trade in Goods and Services – 06-08 October 2021”, Paris: OECD. Retrieved from https://one.oecd.org/document/SDD/CSSP/WPTGS/M(2021)1/en/pdf. - Organisation for Economic Co-operation and Development (OECD) (2022). Revenue Statistics 2022: The impact of COVID-19 on OECD tax revenues, Paris: OECD Publishing. Retrieved from https://www.oecd.org/tax/tax-policy/revenue-statistics-highlights-bro-chure.pdf. - Organisation for Economic Co-operation and Development (OECD) (2023), OECD Handbook on Compiling Digital Supply and Use Tables, Paris: OECD. - Organisation for Economic Co-operation and Development (OECD) and G20 (2021), “International collaboration to end tax avoidance end tax avoidance”. Retrieved from https://www.oecd.org/tax/beps/. - Organisation for Economic Co-operation and Development (OECD), World Trade Organization (WTO) and International Monetary Fund (IMF) (2019), Handbook on Measuring Digital Trade, Version 1, Paris, Geneva and Washington, D.C.: OECD, WTO and IMF. Retrieved from https://www.oecd.org/sdd/its/Handbook-on-Measuring-Digital-Trade.htm. ### National statistical agencies and surveys - Philippine Statistics Authority (2020), “2017 Annual Survey of Philippine Business and Industry (ASPBI) – Economy Wide All Establishments: Final Results” (22 April 2020), Quezon City: Philippine Statistics Authority. Retrieved from https://psa.gov.ph/content/2017-an-nual-survey-philippine-business-and-industry-asp-bi-economy-wide-all-establishments. - Statistics Canada (2017), “The sharing economy in Canada” (28 February 2017), Ottawa: Statistics Canada. Retrieved at https://www150.statcan.gc.ca/n1/daily-quotidien/170228/dq170228b-eng.htm. - Statistics Canada (2019), “The value of data in Canada: Experimental estimates”, Retrieved from https://www150.statcan.gc.ca/n1/pub/13-605-x/2019001/article/00009-eng.htm. - Statistics Canada (2020a), “Canadian Internet Use Survey 2020”, Ottawa: Statistics Canada. Retrieved from https://www23.statcan.gc.ca/imdb/p3Instr.pl?Function=assembleInstr&lang=en&Item_Id=1289522#qb1290266. - Statistics Canada (2020b), “Canada’s services exports through the lens of digital trade”, Ottawa: Statistics Canada. Retrieved from https://www150.statcan.gc.ca/n1/pub/13-605-x/2020001/arti-cle/00005-eng.htm. - Statistics Canada (2021), “Survey of Digital Technology and Internet Use - 2021” (3 August 2021), Ottawa: Statistics Canada. Retrieved from https://www23.statcan.gc.ca/imdb/p3Instr.pl?Function=assembleInstr&lang=en&Item_Id=1317562. - Statistics Canada (2022a), “Gross sales made over the internet by industry and size of enterprise” (13 September 2022), Ottawa: Statistics Canada. Retrieved from https://doi.org/10.25318/2210012401-eng. - Statistics Canada (2022b), “Average expenditure per person on goods and services ordered over the internet by age group” (06 June 2022), Ottawa: Statistics Canada. Retrieved from https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=2210013901 - Statistics Canada (2022c), “Canadian Internet Use Survey 2022”, Ottawa: Statistics Canada. Retrieved from https://www23.statcan.gc.ca/imdb/p3Instr.pl%20?Function=assembleInstr&lang=en&Item_Id=1487379#qb1494610. - Statistics Canada (2022d), “Digital technology and internet use, 2021” (13 September 2022), Ottawa: Statistics Canada. Retrieved from https://www150.statcan.gc.ca/n1/daily-quotidien/220913/dq220913b-eng.htm. - Statistics Canada (2022e), “Trade by exporter and importer characteristics: Services, 2020”, Ottawa: Statistics Canada. Retrieved from https://www150.statcan.gc.ca/n1/en/daily-quotidien/221110/dq221110a-eng.pdf?st=tbgSbNv3. - Statistics Korea (2014), “E-commerce and Cyber Shopping Survey in 2013 and in the Fourth Quarter 2013” (25 February 2014), Daejeon: Statistics Korea. Retrieved from https://kostat.go.kr/board.es?mid=a20101000000&bid=11722&act=view&list_no=313310&tag=&nPage=164&ref_bid=11707,11708,11709,11711,11712,11713,11715,11716,11717,11718,11719,11721,11722,11723,11724,11725,11726,12071,11727,11728,11729,11730,11731,11732,11733,11734,11735,12051,11786,11736,11737,11738,11739,11740,11741,11742,11743,11744,11745,11746,11747,11748,11749,11773,11774,11750,11751,11752,11754,11755,11756,11757,11758,11759,11760,11761,11762,12050,11763,11764,11765,11766,11767,11768,11769,11770,11771,11772&keyField=&keyWord=. - Statistics Singapore (SingStat) (2021), “E-Commerce Revenue of the Services Sector”, Statistics Singapore Newsletter Issue 2, 2021. Retrieved from https://www.singstat.gov.sg/-/media/files/publications/industry/ssn221-pg21-22.ashx. - Visa (2021), Visa Merchant Data Standards Manual – Visa Supplemental Requirements (November 2021). Retrieved from https://usa.visa.com/content/dam/VCOM/download/merchants/visa-merchant-da-ta-standards-manual.pdf. - Tax Administration Jamaica (2022), General Consumption Tax Forms, https://www.jamaicatax.gov.jm/web/guest/general-consumption-tax. - United States Census Bureau (2021), “E-Stats 2019: Measuring the Electronic Economy” (5 August 2021), Report Number E19-ESTATS, Washington, D.C.: United States Census Bureau. Retrieved from https://www.census.gov/library/publications/2021/econ/2019-e-stats.html. ### United Nations, UNCTAD, UNECE, Eurostat, and IMF manuals and reports - United Nations (UN) (2008a), System of National Accounts 2008 (SNA 2008), New York: UN. Retrieved from https://unstats.un.org/unsd/nationalaccount/sna2008.asp. - United Nations (2008b), International Standard Industrial Classification of All Economic Activities, Revision 4, New York: United Nations Statistics Division. Retrieved from https://unstats.un.org/unsd/publication/seriesM/seriesm_4rev4e.pdf. - United Nations (UN) (2011), International Merchandise Trade Statistics: Concepts and Definitions, New York: UN. Retrieved from https://unstats.un.org/unsd/trade/eg-imts/IMTS%202010%20(English).pdf. - United Nations (2015), Central Product Classification (CPC), Version 2.1, New York: United Nations Statistics Division. Retrieved from https://unstats.un.org/unsd/classifications/unsdclassifications/cpcv21.pdf. - United Nations (UN) (2022), “United Nations Statistical Commission, 53rd session”, New York: UN. Retrieved at https://unstats.un.org/unsd/statcom/53rd-session/documents/BG-3v-doc_ISIC-E.pdf. - United Nations, International Monetary Fund (IMF), Organisation for Economic Co-operation and Development (OECD), Statistical Office of the European Union (Eurostat), United Nations Conference on Trade and Development (UNCTAD), World Tourism Organization (UNWTO) and World Trade Organization (WTO) (2010), Manual on Statistics of International Trade in Services 2010 (MSITS 2010), New York, Washington, D.C., Paris, Luxembourg, Geneva and Madrid: UN, IMF, OECD, Eurostat, UNCTAD, UNWTO and WTO. Retrieved from https://unstats.un.org/unsd/publication/seriesm/seriesm_86rev1e.pdf. - United Nations, Eurostat, IMF, OECD, UNCTAD, UNWTO and WTO (2010), Manual on Statistics of International Trade in Services – Compiler’s Guide 2010, New York, Luxembourg, Washington, D.C., Paris, Geneva and Madrid: UN, Eurostat, IMF, OECD, UNCTAD, UNWTO and WTO. Retrieved from https://unstats.un.org/unsd/trade/publications/14-66197-E-MSITS%202010%20Compilers%20Guide_WEB.pdf. - UNCTAD (various): - UNCTAD (2015), “International Trade in ICT services and ICT-enabled services: Proposed Indicators from the Partnership on Measuring ICT for Development – Technical Notes on ICT for Development (No. 3)”. Retrieved from https://unctad.org/en/PublicationsLibrary/tn_unctad_ict4d03_en.pdf. - UNCTAD (2016), “In Search of Cross-Border E-Commerce Trade Data”, UNCTAD Technical Notes on ICT for Development, No. 6, Geneva: UNCTAD. Retrieved from https://unctad.org/en/PublicationsLibrary/tn_unctad_ict4d06_en.pdf. - UNCTAD (2018a), “Updating the partnership definition of ICT goods from HS 2012 to HS 2017. Technical Notes on ICT for Development (No 10)”, Geneva: UNCTAD. Retrieved from https://unctad.org/en/PublicationsLibrary/tn_unctad_ict4d10_en.pdf. - UNCTAD (2018b), “Implementing a Survey on Exports of ICT-enabled Services – Technical Notes on ICT for Development (No. 11)”, Geneva: UNCTAD. Retrieved from https://unctad.org/system/files/official-document/tn_unctad_ict4d11_en.pdf. - UNCTAD (2019), Digital Economy Report 2019: Value creation and capture: Implications for developing countries, Geneva: UNCTAD. Retrieved from https://unctad.org/publication/digital-economy-report-2019. - UNCTAD (2020), Manual for the Production of Statistics on the Digital Economy 2020, https://unctad.org/webflyer/manual-production-statistics-digital-economy-2020. - UNCTAD (2021a), Manual for the Production of Statistics on the Digital Economy 2020, Geneva: UNCTAD. Retrieved from https://unctad.org/webflyer/manual-production-statistics-digital-economy-2020. - UNCTAD (2021b), Digital Economy Report 2021 – Cross-border data flows and development: For whom the data flow, Geneva: UNCTAD. Retrieved from https://unctad.org/publication/digital-economy-report-2021. - UNCTAD (2021c), “Estimates of global e-commerce 2019 and preliminary assessment of covid-19 impact on online retail 2020”, UNCTAD Technical Notes on ICT for Development No. 18, Geneva: UNCTAD. Retrieved from https://unctad.org/system/files/official-document/tn_unctad_ict4d18_en.pdf. - UNCTAD (2021d), Digital Economy Report 2021: Cross-border data flows and development: For whom the data flow, Geneva: UNCTAD. Retrieved from https://unctad.org/publication/digital-economy-report-2021. - UNCTAD (2022a), “COVID-19 boost to e-commerce sustained into 2021”, Geneva: UNCTAD. Retrieved from https://unctad.org/news/covid-19-boost-e-com-merce-sustained-2021-new-unctad-figures-show. - UNCTAD (2022b), “Digital trade: Opportunities and actions for developing countries”, Policy Brief No. 92 (January 2022), Geneva: United Nations. Retrieved from https://unctad.org/system/files/official-document/presspb2021d10_en.pdf. - UNCTAD (2022c), Digitalization of Services: What does it imply for trade and development?, Geneva: UNCTAD. Retrieved from https://unctad.org/system/files/official-document/ditctncd2021d2_en.pdf. - UNCTAD (2022d), E-commerce and the digital economy in LDCs: At breaking point in COVID-19 times, Geneva: UNCTAD. Retrieved from https://unctad.org/publication/e-commerce-and-digital-econ-omy-ldcs-breaking-point-covid-19-times. - UNCTAD (2022e), “Supporting countries to measure the digital economy for development” (02 December 2022), Geneva: United Nations. Retrieved from https://unctad.org/news/supporting-countries-measure-digi-tal-economy-development. - UNCTAD (2023), Measuring the value of e-commerce, Geneva: UNCTAD. Retrieved from https://unctad.org/publication/measuring-value-e-commerce. - UNECE, Eurostat and OECD (2011), The Impact of Globalization on National Accounts, Geneva, Luxembourg and Paris: UNECE, Eurostat and OECD. Retrieved from https://unece.org/DAM/stats/publications/Guide_on_Impact_of_globalization_on_national_accounts__web_.pdf. - United Nations Statistics Division (UNSD) (2022), “DZ. 6 – Recording of data in the National Accounts”, New York: UNSD. Retrieved from https://unstats.un.org/unsd/nationalaccount/RAdocs/DZ6_GN_Recording_of_Data_in_NA.pdf. ### WTO, WCO and trade-focused datasets and manuals - World Customs Organization (WCO) (2022), “Framework of Standards on Cross-Border E-Commerce”, Brussels: WCO. Retrieved from http://www.wcoomd.org/-/media/wco/public/global/pdf/topics/facilitation/activities-and-programmes/ecommerce/wco-framework-of-standards-on-crossborder-ecom-merce_en.pdf?la=en. - World Trade Organization (WTO) (1998a), “Work Programme on Electronic Commerce”, Geneva: WTO. Retrieved from https://docs.wto.org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/L/274.pdf&Open=True. - World Trade Organization (WTO) (1998b), “Declaration on Global Electronic Commerce”, Geneva: WTO. Retrieved from https://docs.wto.org/dol2fe/Pages/SS/directdoc.aspx?filename=Q:/WT/MIN98/DEC2.pdf&Open=True. - World Trade Organization (WTO) (2019), Joint Statement on Electronic Commerce, Geneva: WTO. Retrieved from https://docs.wto.org/dol2fe/Pages/SS/directdoc.aspx?filename=q:/WT/L/1056.pdf&Open=True. - World Trade Organization (WTO) (2020), “The General Agreement on Trade in Services (GATS): objectives, coverage and disciplines”, Geneva: WTO. Retrieved from https://www.wto.org/english/tratop_e/serv_e/gatsqa_e.htm. - World Trade Organization (WTO) (2022), “Highlights of world trade in 2021”, in World Trade Statistical Review 2022, Geneva: WTO. Retrieved from https://www.wto.org/english/res_e/booksp_e/wtsr_2022_c2_e.pdf. - World Trade Organization (WTO) (2023), “Global Trade Outlook and Statistics”, Geneva: WTO. Retrieved from https://www.wto.org/english/res_e/booksp_e/trade_outlook23_e.pdf. - World Trade Organization (WTO) and United Nations Conference on Trade and Development (UNCTAD) (2022), “International trade in digitally-deliverable services, value, shares and growth, annual”, Geneva: WTO and UNCTAD. Retrieved from https://unctadstat.unctad.org/wds/TableViewer/tableView.aspx?ReportId=158358. - Wettstein, S., Liberatore, A., Magdeleine, J. and Maurer, A. (2019), “A global trade in services data set by sector and by mode of supply (TiSMoS)”, Geneva: WTO. Retrieved from https://www.wto.org/english/res_e/statis_e/daily_update_e/Tismos_methodology.pdf. ### Research papers, working papers and technical contributions - Rassier, D. G., Kornfeld, R. J. and Strassner, E. H. (2019), “Treatment of Data in National Accounts. Paper prepared for the BEA Advisory Committee”, Paper prepared for the Bureau of Economic Analysis (BEA) Advisory Committee. Retrieved from https://www.bea.gov/system/files/2019-05/Paper-on-Treatment-of-Data-BEA-ACM.pdf. - Santiago Calderón, J. and Rassier, D. G. (2022), “Valuing the U.S. Data Economy Using Machine Learning and Online Job Postings”, in Basu, S., Eldridge, L., Haltiwanger, J. and Strassner, E. (forthcoming), Technology, Productivity, and Economic Growth, Chicago: University of Chicago Press. Retrieved from https://www.nber.org/system/files/chapters/c14739/c14739.pdf. - Shirono, K., Das, B., Fan, Y., Chhabra, E. and Carcel-Villanova, H. (2021), “Is Mobile Money Part of Money? Understanding the Trends and Measurement”, IMF Working Paper WP/21/177, Washington, D.C.: International Monetary Fund (IMF). Retrieved from https://www.imf.org/-/media/Files/Publications/WP/2021/English/wpiea2021177-print-pdf.ashx. - Sorbe, S., Gal, P., Nicoletti, G. and Timiliotis, C. (2019), “Digital Dividend: Policies to Harness the Productivity Potential of Digital Technologies”, OECD Economic Policy Papers, No. 26, Paris: OECD Publishing. Retrieved from https://doi.org/10.1787/273176bc-en. - Statistics Canada entries on experimental estimates and surveys (see National statistical agencies section above). *Handbook on Measuring Digital Trade — Bibliography* ### BIBLIOGRAPHY ### BIBLIOGRAPHY ### Acronyms and abbreviations - AEG UN Advisory Expert Group on National Accounts - AI artificial intelligence - ASYCUDA UNCTAD Automated System for Customs Data - B2B Business-to Business - B2C Business-to-Consumer - B2G Business-to-Government - BEA US Bureau of Economic Analysis - BOPCOM IMF Committee on Balance of Payments Statistics - BPM6 Balance of Payments and International Investment Position Manual. Sixth Edition - BPO Business Process Outsourcing - BOJ Bank of Jamaica - C2C Consumer-to-Consumer - CBEC Cross-Border e-Commerce - CIF Cost, Insurance and Freight - CPA EU classification of products by activity - CPC Central Product Classification - DEA Digital Economy Agreement - DIP Digital Intermediation Platform - DSTRI OECD Digital Services Trade Restrictiveness Index - EBOPS Extended Balance of Payments Services Classification - ETCD Electronic Trade Customs Declaration - EDI Electronic Data Interchange - EGATUR Tourism Expenditure Survey - EU European Union - FATS Foreign Affiliates Statistics - FOB Free on Board - FRONTUR Tourist Movements at Borders - GATS General Agreement on Trade in Services - HBS Household Budget Survey - HS Harmonized Commodity Description and Coding System - ICT Information and Communications Technology - ICT-E Annual Information and Communications Technology usage and e-commerce in Enterprises survey - ICT-H Survey on Equipment and Use of Information and Communication Technologies in Households - IMTS International Merchandise Trade Statistics: Concepts and Definitions - INE Instituto Nacional de Estadística - ISIC International Standard Industrial Classification of All Economic Activities - IT Information Technology - ITO Information Technology Outsourcing - ITRS International Transaction Reporting System - ITS International Trade in Services - ITSS International Trade in Services Statistics - KPO Knowledge Process Outsourcing - LCU Large Case Unit - LDC Least-Developed Country - MNE Multinational Enterprise - MNO Mobile Network Operator - MOSS VAT Mini One Stop Shop (data collection scheme) - MSME Micro, Small and Medium-sized Enterprise - n.d. no date - NACE Nomenclature of Economic Activities - NFT Non-Fungible Token - n.i.e. not included elsewhere - NPISH Non-Profit Institutions Serving Households - ONS UK Office for National Statistics - OSS One Stop Shop (data collection scheme) - R&D Research and Development - RTA Regional Trade Agreement - RTS Residents Travel Survey - SNA System of National Accounts - SPTS Statistics on Products in the Trade Sector - STATIN Jamaican Statistical Institute - SUTs Supply and Use Tables - TAJ Tax Authority Jamaica - TiSMoS WTO Trade in Services by Mode of Supply - TT-ISIC Task Team on International Standard Industrial Classification of All Economic Activities - UN United Nations - UNCEISC United Nations Committee of Experts on International Statistical Classifications - VAT Value-Added Tax - VoIP Voice over Internet Protocol - WCO World Customs Organization - YoY Year-on-Year ### Publication and contact details (WTO Handbook on Measuring Digital Trade) - WTO ISBN 978-92-870-7360-0 (print) - WTO ISBN 978-92-870-7359-4 (PDF) - The report is also available in Arabic, French and Spanish - Copies can be ordered from: WTO Online Bookshop http://onlinebookshop.wto.org - World Trade Organization - Centre William Rappard - Rue de Lausanne 154 - CH-1211 Geneva 2 - Switzerland - Tel. switchboard: +41 (0)22 739 51 11 - email: enquiries@wto.org - www.wto.org ### What is “digital trade” — conceptual focus and measurement criteria - Digital technologies enable: - buyers and sellers to place and receive orders on a global scale - instantaneous remote delivery of services directly into businesses and homes, including internationally - The Handbook on Measuring Digital Trade focuses on two criteria for defining digital trade: - digital ordering across borders - digital delivery across borders - The Handbook sets out a conceptual and measurement framework for digital trade that aligns with broader standards for macroeconomic statistics. ### Purpose, scope, and support for statistical compilers - Aims: - Help statistical compilers address policymakers’ needs for statistical evidence on digital trade - Content and tools: - Extensive compilation guidance drawing upon substantive inputs and case studies from developed and developing economies - Coverage of a variety of survey and non-survey sources - A proposed reporting template to support compilers in the production and dissemination of digital trade statistics - Outcome: - Establishes a comprehensive agreed approach and a valuable shared foundation for understanding and measuring digital trade in an internationally comparable way - Provides a resource for technical assistance and statistical capacity-building by the four co-authoring partner organizations ### Partnership and authorship - Second edition outcome of a partnership between: - International Monetary Fund (IMF) - Organisation for Economic Co-operation and Development (OECD) - United Nations Conference on Trade and Development (UNCTAD) - World Trade Organization (WTO) *Source: BIBLIOGRAPHY, Handbook on Measuring Digital Trade (excerpts).* --- _Source: https://www.imf.org/-/media/files/publications/books/2023/english/hmdtea.pdf_