Mirror data asymmetry in international trade by commodity group: the case of intra-Community trade
Mirror data asymmetry in international trade by commodity group: the case of intra-Community trade

Summary/Abstract: Research background: Transactions in international trade of goods are recorded in two sources, on the side of the seller's country and on the side of the buyer's country. The confrontation of such data makes it possible to measure their quality. An inconsistency between the data is called mirror data asymmetry. Purpose of the article: The aim of the paper is to adapt the methods developed by the Authors to study mirror data asymmetry to commodity group markets examination. The quality of data on trade within specific commodity groups (CN chapters) in intra-Community trade was compared. The data were aggregated by country. The indicators used allow for the indication of commodity groups with high mirror data compatibility and those with data asymmetry between intra-Community supplies (ICS) and acquisitions (ICA). Moreover, the commodity groups for which the value-based and quantity-based approaches give different results have been identified. Methods: Based on the literature on the subject and their own research, the Authors have developed a group of methods for studying the asymmetry of mirror data. The proposed indicator formulas are based on various data aggregation approaches. The research used data on intra-Community supplies and acquisitions of goods broken down into 97 chapters of the Combined Nomenclature (CN). Differences between the ICS and ICA in particular commodity groups were aggregated for all pairs of EU countries. The data comes from the Comext database, provided by Eurostat.Findings & value added: The results of the analysis are rankings of the Combined Nomenclature (CN) chapters by the quality of data on ICS and ICA. Lists of CN chapters have been created for discrepancies both in value and weight of goods. Thus, areas of necessary intensification of the work of public statistics services to improve data reliability were identified.

• Details
• Contents