22-24 November 2004 16 November 2004

IMPLEMENTATION OF STATISTICAL RESEARCH AGENDA:

RECONCILIATION OF MERCHANDISE TRADE

1. Introduction

As mentioned in Paper SCCS/2004/29/11, the CARICOM Secretariat has been compiling detailed regional Merchandise Trade database since 1990. During the process of establishing the regional database, often queries on perceived anomalies in the data are submitted to Member States. Additionally, the analysing of data on intra-regional trade often shows inconsistencies in the corresponding imports and exports among Member States. As a consequence of these experiences, the Secretariat has included in its Work Programme, the reconciliation of trade data aimed at improving the quality of the data of Member States and also to achieve greater harmonisation of the data. This Paper therefore attempts to establish an approach to reconciling the intra-regional trade data through identifying some of the major reasons why the published figures of two countries may differ. It is part of a process to initiate in the Statistics Sub-Programme increase focus on statistical research and analysis.

2. Approach to Reconciling Trade Data

Generally, trade between two countries, A and B, should be a mirror image of each other. That is, exports of Country A to Country B should be the same as Country B imports from Country A. It is often the case that statistics produced by countries on their Merchandise Trade with the rest of the world differs from the statistics published by their trading partners. These differences reflect both legitimate conceptual differences between import and export statistics as well as genuine errors or discrepancies in the data.

A reconciliation study can be undertaken to investigate the causes of the differences in the trade statistics between countries. It is important to note that the results of this study do not constitute revisions to any country's official statistics; however, they may help to improve the understanding of the trade statistics of the respective countries and to the production of a more harmonised trade data across the Region.

In the short term therefore, this exercise aims to identify, explain and assess the causes of discrepancies between the trade data of two countries of the Region. The long-term objective can be viewed as the harmonisation of the conceptual frameworks of Member States. The reconciliation exercise will improve the quality of the data through the revealing of measurement errors and gaps, providing trade negotiators and analysts with more reliable and consistent data. This exercise could also lead to the development of new procedures in the compilation of foreign trade statistics.

3. Choice of Data Series

For this exercise, import or export data can be used as the benchmark for the purposes of comparison between Member States. However, it is open to debate as to which data series imports or exports, is the more accurate one in the case of the CARICOM Region. The reporting of import data may be viewed as more accurate because they are reported in sufficient detail to allow Customs administration to apply duties, taxes or other regulatory controls. However, there are problems that may affect the quality of the import data such as inaccurate recording of values and quantities in the Customs invoices with the aim of impacting on the duties that are applied. On the other hand, the narrow band of traditional products such as Sugar, Petroleum, Bananas and so forth may be reported with a greater degree of accuracy with the actual value of exports often being obtained from the records of the exporting companies/boards.

For the Member States of CARICOM, this exercise can provide each Member State with a better understanding and a more accurate reflection of their corresponding trade flows leading to better quality data and insights into methods of improving future data collection and/or production of merchandise trade statistics.

4. Some of the Major Reasons for Disparities in the Merchandise Trade Data between Countries

(i) Trade System

There are two recognized systems for recording Trade: the general trade system and the special trade system.

General Trade System: Under this system goods are recorded as they enter or leave the country. For imports, this system includes goods imported into free trade zones or customs warehouses, regardless of whether the goods are intended for home use or re-export. Similarly, goods exported from free trade zones or customs warehouses are recorded at the time they leave the country.

Special Trade System: Under this system goods are recorded only when they enter into free circulation or for inward processing in the country of receipt. Therefore, goods entering into the free trade zone or customs warehouse are excluded from trade at the time of importation but are subsequently recorded at the time of release to home use or inward processing. Similarly, outgoing goods from customs warehouses are not recorded as exports.

As a result, countries may capture their trade transactions using the General or Special trade systems. Since goods are classified differently under these two major trade systems then this will create discrepancies when trying to reconcile data for two countries using different trade system. Both trade systems are used within CARICOM.

(ii) Reporting Time Lags

Time differences in recording transaction will affect data comparability. For example, goods shipped towards the end of the reporting period in an exporting country may arrive in the next period in the importing country. Thus, the goods will be recorded in two different periods in the exporting and importing countries.

Also there will be time lags with countries using the special trade system, since the system only records the goods when it enters the economy of the receiving country. Goods stored in bonded warehouses are not recorded until removed from the warehouse. This approach may increase timing differences and will affect data comparability.

In any trade reconciliation exercise, time lags must not be overlooked, it may contribute to one of the major reasons to explain discrepancies.

(iii) Geographical coverage

The statistical territory adopted by the countries may differ from one country to another. For example a country may group more than one statistical territory together while the corresponding trading partner may report it as separate trading countries. Thus there may be differences in geographic coverage, which may create significant discrepancies.

In any reconciliation exercise it is important to have a clear list of statistical territories of the two countries for which the exercise is undertaken.

(iv) Confidentiality provisions

Confidentiality can affect the reporting of information by products or trading partners. The application of secrecy rules should only affect figures at a detailed level. However, goods may be reclassified for confidentiality reasons. Country confidentiality occurs where figures may be assigned under fictitious country codes, product confidentiality is reflected in the disguising of products under superior or fictitious headings and total confidentiality is complete exclusion of items from the trade statistics, or the aggregation in an unallocated category.

Confidentiality, provisions can therefore contribute to differences between trading partners.

(v) Exchange rate

The value of a transaction is recorded in the country's national currency, even if the transaction was completed in another currency. For comparison, trade data must be converted into the same currency. The use of different exchange rates can result in discrepancies when the data of the two countries are compared particularly in instances where there are wide or numerous fluctuations in the official exchange rate to the foreign currency being used for the conversion process.

(vi) Indirect Trade

This situation may be an important source of discrepancy in a reconciliation exercise. Very often there are situations where goods are shipped to a country via an intermediate third country. The exporters record the country of destination known at that time without realising the final destination of the goods is in fact another country.

(vii) Low Value Trade

Some countries do not include in their trade, transactions below a certain value while other countries may include the value of the trade regardless of the amount. This affects data comparability. Also it may be the case that a country may report small transactions under a common product code regardless of the commodity traded. This will also affect comparability at a more detailed level.

(viii) Under-Over Reporting

There are several reasons leading to the under or over valuation of goods traded. In most cases they are due to non-reporting, fraudulent declarations, reporting errors, irregular trade, repairs and maintenance. Also specific goods may be defined differently from country to country. For example, trade in gold coins, military goods etc. may be recorded as ordinary trade and tabulated accordingly whereas it can also be treated as differently and recorded separately from ordinary trade.

(ix) Misclassification

Despite the use of the Harmonised System (HS) by many trading countries for the processing of their trade data, there are still differences in interpreting and applying these codes. As a result, commodities may be misclassified based on the interpretation of the person examining the commodities. This misclassification will result in differences when reconciling at the detailed level.

(x) CIF and FOB Conversion

Generally exports are recorded FOB (free on board) at the frontier of the exporting country, while imports are recorded CIF (cost, insurance and freight) at the frontier of the importing country. To compare data, it will be necessary to have the import and export data on the same basis (i.e. CIF or FOB). It is more common to adjust the data to the FOB basis rather than the CIF since the freight and insurance is known.

A brief look at Barbados and Trinidad and Tobago's trade with each other

A brief look at the statistics of Barbados and Trinidad and Tobago for 2003 revealed that the total discrepancy between Barbados's exports statistics and Trinidad and Tobago's imports statistics was EC$16.2 million. On the other hand, Trinidad and Tobago's exports statistics differ from Barbados' imports statistics by EC$84.6 million. Attached is a Table outlining the discrepancies.

One of the major reasons for the difference in the figures is the trade system used by the two Member States. Barbados records its trade using the general trade system while Trinidad and Tobago uses the special trade system. This will result in discrepancies since the trade system used affects the time at which the goods are recorded. However Barbados has been maintaining a special trade system for comparison purposes, which will assist greatly in a reconciliation exercise.

Another possible reason may be due to exchange rates in use. Since the two Member States uses different currencies to record their trade converting to a common currency for comparison may result in discrepancies.

Also both Member States record their imports at CIF and their exports at FOB. Thus, comparison of the two figures will definitely be different.

These are only some of the reasons why the trade data may differ and further analysis will be required to identify the sources of the difference in the trade figures of both countries.

ACTION REQUIRED

The Meeting is invited to:

(i) note the developmental work being undertaken by the Secretariat in the area of Merchandise Trade in goods as part of an wider statistical research agenda to be implemented;

(ii) consider the significance of this reconciliation relative to improved quality and harmonisation of trade statistics.

*****

ATTACHMENT

Table 1: Discrepancies between Barbados' and
Trinidad and Tobago's Statistics - 2003

	EC$000
Barbados' Exports to Trinidad and Tobago	76,570
Trinidad and Tobago's Imports from Barbados	60,357
Difference	16,213

Trinidad and Tobago's Exports to Barbados	553,745
Barbados' Imports from Trinidad and Tobago	638,389
Difference	(84,644)