The quality of trade data is essential for any empirical study that uses them to construct a variable representing trade.
| The quality of trade data is essential for any empirical study that uses them to construct a variable representing trade. However, previous studies have largely ignored the issue of trade data discrepancy. This discrepancy occurs when a country's claim on sectoral bilateral trade is different from the value reported by its trade partner. Using a size ratio and correlation coefficient computed from the values reported by both parties of bilateral trades, this study examines trade data discrepancy in 20 trade sectors from 1987 to 2016 between Malaysia and its three major Asian trade partners, i.e. China, Singapore and Japan. We detect alarming discrepancy in Malaysia's trade data. We also find that this discrepancy depends on the trade partner, time and trade sector. This study calls for attention of Malaysian policymakers regarding the possibility of customs mismanagement. We recommend that future empirical studies utilising trade in the analysis to either reconcile the data or to deploy both data sources in the econometric analysis as a robustness check. |
