Abstract

In the European Union, National Statistical Institutes, National Banks, Eurostat and the ECB have been working for several years to reduce trade asymmetries, for example, through dedicated workshops, specific meetings, and methodological advice. Notwithstanding the important progress being made, trade asymmetries still exist which makes it hard for practitioners and researchers to build macroeconomic models and assess detailed economic relationships between countries.
For the construction of FIGARO (Full International and Global Accounts for Research in input-Output analysis), an approach was specifically developed to eliminate the trade asymmetries, which uses estimation methods and exploits all information available for trade in goods and services in trade statistics but also in national accounts.
As regards the goods, a combination of the EU Comext and UN Comtrade databases is used. Data are recoded and combined into a unified data structure, and non-allocated trade is imputed whenever this is possible using mirror data. Remaining asymmetry issues are solved through a balancing process, based on a symmetry index (or reliability index). The balanced trade flow is then broken down into 3 categories: quasi-transit trade (Q), domestic trade (D) and re-exports (R). The QDR methodology is crucial for building inter-country input-output tables as it makes possible to break down the export vector of the national accounts’ use table of domestic input by partner country. Finally, adjustments for goods sent abroad for processing and merchanting activities are done to align the concepts underlying the balanced bilateral trade estimates with the concepts and coverage of the system of national accounts.
For the services, data confidentiality and the use of different data sources and classifications (EBOPS versus CPA/CPC) can complicate the scheme. Once a complete (albeit unbalanced) dataset of bilateral trade flows of services data is achieved, the same balancing approach as for the goods (symmetry index) is followed to estimate a single value for each bilateral trade flow.
The different steps of the FIGARO methodology to overcome the trade asymmetries are being continuously monitored during the process in order to ensure high quality of the obtained results. The monitoring includes an assessment of the changes made in the complete processed raw data for trade in goods and services by EU Member State and by CPA code. Shares of quasi-transit trade (Q), domestic trade (D) and re-exports (R) data for trade in goods, as well as the extent of the adjustments for goods sent abroad for processing and merchanting activities, for each EU Member State and CPA code are also closely monitored during the process.
The presentation, after explaining the main methodological steps used in FIGARO to mitigate trade asymmetries, will show the different results of the quality assessment, including recommendations for future collaboration with other Eurostat´s units and Member States involved in the dedicated workshops and other meetings to reduce trade asymmetries.