Presentation on theme: "An alternative presentation of trade by Product and Industry. Checking results against survey data OECD Statistics Directorate. Bettina Wistrom, Alena."— Presentation transcript:
An alternative presentation of trade by Product and Industry. Checking results against survey data OECD Statistics Directorate. Bettina Wistrom, Alena Brin.
Outline Background. Why is this useful? Presentation of the correlation tables. New features of the project Limits of present exercise. Comparing export of services by industry with survey data.
Background At the 2010 WPTGS, a note providing an estimated example on how to combine conceptually exports of goods and exports of services in one Product and Industry presentation using existing correspondences was presented. The example was based on actual export data extracted from OECD databases for year 2007 using the links of principal products to industry sections. The allocation assumptions were made explicit.
Two separate frameworks for detailed trade in goods and services. No existing structure that presents detailed trade in goods and services (i.e. trade by products) together in the same unique framework. The unified framework is here the Industry (ISIC rev 3.1) classification using the product classification CPC rev1.1 as a link as far as possible to HS 2002 and EBOPS 2002.
Different conceptual bases. All the detailed trade in goods is presented following HS (about 5000 items at 6 digit level) and all the detailed trade in services is presented following the Extended Balance of Payments Services Classification (about 85 items in EBOPS 2002) within the BOP framework. The two systems have a different conceptual basis and they cannot be combined without further adjustments.
Why a proposal for an alternative presentation of trade by product. The need for relating trade statistics to SUT and IO is an argument for producing integrated statistics on trade in product by activity This could also improve analytical links to production, employment, enterprise, AMN, and direct investment statistics.
Preparation of correspondence The correspondence was established on the merchandise side, between ISIC rev. 3.1 and HS 2002 using CPC 1.1 as an intermediary link, via UNSD correspondence tables. On the service side, the correlation table used is an elaboration of the ISIC Categories for Foreign Affiliates (ICFA)– EBOPS 2002 correspondence table as presented in the annex 4 of the Manual on Statistics of International Trade in Services 2002 using CPC 1.1 as a link as far as possible.
Summary correspondence table 1/3 Half to value of Advertising market research and public opinion polling (278) Allocated to ISIC D22 publishing and printing Half to value of Other business services(284) Split between ISIC D22 publishing and printing and ISIC D 28, manufacture fabricated metal products
Summary correspondence table 2/3 Waste treatment and Depollution allocated to D37 recycling Underestimation of exports of services by the manufacturing Industry!
New features. The product by industry exports series have been computed using a SAS program and are prepared for data ranging from 2002 to the latest period for OECD countries and BRICS (still with partner world only). An elaboration of methodology presented last year is that both in the goods and the service correspondences, when there is a multiple correspondence of the HS 2002 item or the EBOPS 2002 item to several ISIC rev. 3.1 sections, the export value of the product is split equally across the ISIC rev. 3.1 sections as was the case only for services previously.
Splitting HS items across ISIC sections (France)
Limits to the exercise Some of the main challenges of the implementation of the revised nomenclatures (and in particular, goods for processing moving from goods to services in BPM6 under the item manufacturing services provided on inputs owned by others) are not dealt with. No attempt to estimate neither the service element possibly included under merchandise trade in CD roms (in transactions of non customized software, provided on disks with a perpetual license to use) nor the size of distribution services included in merchandise trade.
Export data as no CIF-FOB adjustment needed Experiment not extended to imports. –The link between industry and import of products is too loose as a specific industry is likely to import any type of product. –There is also a better match between the merchandise trade (customs based) and the service trade (from balance of payments frameworks) on the export side as no CIF- FOB adjustment needs to be performed.
Comparing results with survey data Results of the conceptual correspondence on the service side are compared with survey data obtained by Germany, Austria and Canada.
Largest service exporting industry (Germany) Rank 1234567 Micro dataD:33%I: 26%K:17%J:16%G: 5%F:2% Conceptual Correspond -ence D: 2%I:34%K: 26%J:8%H:8%G:7%F: 6% D: Manufacturing I: Transport, storage and communications K: Real estate renting and business activities J: Financial Intermediation G: Wholesale and retail trade; repairs of motor vehicles, motorcycles and personal and household goods H: Hotels and restaurants F: Construction
Exports of goods and services by industry- Germany
Comparing results for Austria ISIC K: 17% ISIC I: 44%
Largest service exporting industry (Austria) Rank 1234567 Micro dataI: 44%K:17%D:17%G: 15%F:4% Conceptual Correspon- dence I:44%K: 20%D: 1%H:18%G:8%J:5%F:3% D: Manufacturing I: Transport, storage and communications K: Real estate renting and business activities J: Financial Intermediation G: Wholesale and retail trade; repairs of motor vehicles, motorcycles and personal and household goods H: Hotels and restaurants F: Construction
Exports of goods and services by industry- Austria
Comparing results for Canada ISIC K: 28% ISIC I and G: 27%
Canada: Conceptual correspondence ISIC I and G: 31% ISIC K: 30%
Largest service exporting industry (Canada) Rank 12345 Micro dataK: 28%I+G: 27%J:15%D:9% Conceptual Correspon- dence K:30%I+G:31%J:11%D: 3%H:9% D: Manufacturing I: Transport, storage and communications K: Real estate renting and business activities J: Financial Intermediation G: Wholesale and retail trade; repairs of motor vehicles, motorcycles and personal and household goods H: Hotels and restaurants F: Construction
Exports of goods and services by industry- Canada
Conclusion First comparisons between the trade in services allocation that results from the conceptual allocation are relatively encouraging, with the exception of manufacturing services. The next step would be to extend these comparisons to other countries that also produce survey based trade in services by industry and start estimating distribution services if countries have started some work on this issue
Questions to delegates Other countries producing survey based trade in services by industry? Any estimation available of distribution services embedded in commodity exports? Can this presentation of goods and services go (partially) live on OECD.stat for delegates to examine and comment?