Agenda item 9 Impact of Globalization on national

Agenda item 9 Impact of Globalization on national Accounts: E-Commerce Mark Vancauteren, Hasselt University and Statistics Netherlands Group of Experts on the Impact of Globalisation on National Accounts (GGNA), 28 -29 April 2010.

Outline of the paper • • E-Commerce: definition(s) Economic Rationalee Statistical Treatment in International Standards Measurement – Classification, identification and coverage – Additional Data Challenges • E-commerce within Statistics Netherlands

Definition(s) • No commonly accepted definition: (international) E-commerce involves within (cross) border transactions through the internet or other external networks. • Definition on e-commerce varies because of three major issues: – Framework – Sellers and buyers: B-to-C; B-to-B, C-to-C – Type of electronic exchange: e. g. internet versus other electronic networks

Sources: U. S. Bureau of the Censusb, OECDc, a. GAO (2002), d. UNCTAD

• New definition (OECD Working Group efforts started in 2009): – “…to the sale or purchase of goods and services conducted over computer networks by methods specifically designed for the purpose of receiving or placing orders. The goods or services are ordered by those methods, but the payment and the ultimate delivery of the goods and services do not have to be conducted online…”

Economic Rationale • E-commerce affects various components of National Accounts: “a process that mediates transactions (trade) of selling goods & services through electronic exchange” – cost reductions – more competition – better re-organization of production processes

Statistical Treatment • Statistical guidance on e-commerce in international manuals is yet limited: – Electronic of supply of products – Residence of supplier • BPM 6: – physical deliveries: the usual statistical treatment is applied to goods and services – Electronic delivery (digitized products): charges are included in services – Allocation of electronic products to EBOP components

• GATS and mode of supply: – Mode 1: cross border trade – Mode 2: consumption abroad • Task force on the Ro. W (BPM Working Group) recommends “Information from credit card operations”

Measurement • Classification of on-line deliveries for services and digitized goods (with physical counterparts) • Examples include: online payments for travel arrangements (hotel and travel), insurance purchases, software and other related services • Differences exist in the practicability and quality of digitized substitutes of physical goods/services (e. g. , book) • GATT/GATS rules

• Problems with cross-border e-commerce: – Low export values: one-time low valued purchases (Bto-C commerce) – Underreported transactions of small-scale services – New services: e-commerce is another channel to reach consumers (B-to-C, B-to-B) – Residents: Intrastat, Ro. W

• Problems with cross-border e-commerce: – Discrepancy between the retail consumption expenditure within national borders and the gross retail index • The turnover index does not take into account domestic sales when the invoice is issued from abroad; although it included exports from the domestic country (nondomestic orders) => retail trade (annex C) and other services (annex D) does not require this distinction • Implications of the emergence of small retail trade of ecommerce services of electronically delivered products, with domestic and foreign turnover – Same problem occurs with the calculation of output prices (for the calculation of PPI’s)

• Is internet trade sufficiently important to be included in the CPI index: – Weight of e-commerce in spending in total consumer expenditure: e. g. , for the NL, 2 -3% of total expenditure per household) – Lower prices? e. g. , for the NL, about 50% of respondents find the price advantage vis-à-vis traditional retail trade prices an important determinant for on-line purchases; higher PPP through VAT and exchange rate differences)

– Methodological account of e-commerce for substitution-, quality-, new goods- and outlet substitution bias • Switch from traditional outlets towards wholesalers and internet firms; wider availability of used goods and services though C-to-C commerce and auction sales; new goods from abroad ((e. g. i. Phone) – Effect of lower retailing prices of substitutes? • Finkel and Yiftach, 2003: price on internet market are more volatile than the retailing market for a similar basket of goods • Brynjolfsson and Smith, 2000: price dispersion is higher among traditional retail outlets

Experience Statistics Netherlands • NACE code 6312 “web portals”; code 4791 “retail sale via mail order houses or via internet” • GDP distinction between e-commerce and other sales channels is not made. Is this possible? • Majority of VA to e-commerce is captured in the statistics including online sales abroad with the exception of purchases from foreign websites • Underestimation of C-to-C; services from foreign internet facilities should be recorded in NA

• Survey: Automation and ICT (2000 -2008) – 2008: • % of turnover generated via a public website • % of turnover generated via automated data exchange or other private systems • Total e-turnover: the Netherlands, EU, non-EU • % of purchases generated via electronic networks • Total e-purchases: the Netherlands, EU, non-EU • Definition (Eurostat, OECD): "Placing or receiving orders for goods or services through electronic networks, regardless (!) of delivery and payment methods. Excluding orders by telephone, fax, or conventional e-mail. "

• Business statistics: • The sample does not capture e-commerce sales through the internet in case such activities are considered as secondary (side) activities. • The response on the question on e-commerce is often inaccurate and there is less control for imputation. • Structural mistakes: for instance, retailers that moved to e-commerce sales as a primary activity are still being classified as retail sales. • The interpretation of the e-commerce definition is often misunderstood. • Imports and exports of e-commerce goods and services are not observed.

Conclusion • OECD has proposed a new definition which has the potential to solve many of the difficulties related to the functionality and supply of data information. • The statistical challenges on measurement issues and implications on different areas of national accounts (international trade, CPI, PPI, transportation margins) are also discussed. • E-commerce under form of products ordered and paid through the internet poses challenges when it is aligned to the transaction mode of physical versus electronic delivery.

• At this moment, the statistical guidelines on e-commerce in international manuals are very limited. • Regarding classification, the focus has been more on the set of products that are delivered electronically however some work remains to be done in classifying these ecommerce products and services more explicitly. If an ecommerce or some variant is established in the ISIC then consideration should also be given at least to a corresponding division in the EBOP with clear distinction between physical and electronic delivery.