Munich Personal RePEc Archive

Reconciling International Trade Data

Shaar, Karam (2017): Reconciling International Trade Data.

[img]
Preview
PDF
MPRA_paper_81572.pdf

Download (630kB) | Preview

Abstract

International trade data are in substantial measurement error. Data reported by some countries mean next to nothing. This study develops an index of trade data quality based on the consistency between a country’s claims on bilateral trade and the corresponding claims of the rest of the world from 1962 to 2014. The index takes the relative significance of each partner and data availability into account. We produce a more reliable set of bilateral and total international trade data using the index. Findings include (a) the actual exports of most countries with low data quality are considerably higher than self-reported. (b) Corruption and poor data quality are strongly correlated. (c) Global trade data quality has been deteriorating in the past three decades even though more countries have improved their data quality over time. This is because low-quality reporters have recently increased their share in global trade. (d) China tends to under-report exports and over-report imports. (e) There is only a trivial difference between US self-reported and reconciled data. The same applies to all high-quality reporters. We recommend future studies on trade use our reconciled data.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.