Methodology

Data sources

• Statistics Canada, International Trade Division
• Statistical Department of the Customs General Administration of China
• Census and Statistics Department of Hong Kong

Trade data adjustment

Indirect trade adjustment
Control adjustment
Price mark-up adjustment
Re-export adjustment
Valuation adjustment
Residual adjustment

Indirect trade adjustment

Westbound

Canada does not keep separate records for goods shipped directly (direct exports) and goods shipped via intermediary economies to a partner country (indirect exports). However, Chinese import statistics are published by country of origin (COO) and by country of consignment (COC), the county or region from which the goods are dispatched to China, consequently indirect trade can be estimated using Chinese data.

Indirect trade was estimated for goods travelling via Hong Kong, via the United States and via Other economies, which included all the intermediary economies other than Hong Kong and the United States.

Estimates were based on Chinese data net insurance and freight costs using values for transactions that showed Canada (CA) as the country of origin (COO equals CA) but not as the country of consignment (COC does not equal CA).

The values for goods travelling via Hong Kong were estimated by totalling the values of Canadian import data where the country of origin was China (COO equals CN) and where the country of export was Hong Kong (COE equals HK).

Eastbound

In the opposite flow, Canadian data includes both COO and COE. Therefore, based on Canadian import data, values for indirect trade via the United States were estimated by totalling the values of Canadian import data where the county of origin was China (COO equals CN) and the country of export was the United States (COE equals US).

The values for indirect trade via 'Other' were estimated using the import data where the country of origin was China but the country of export was neither China, nor Hong Kong, nor the United States.

For adjustment purposes these estimates were subtracted from the published Chinese import data.

The differences between the Canadian indirect imports and the respective Chinese reported indirect exports were used to align the published data. The differences between the two figures were taken for adjustment purposes in order to reduce the incidence of double counting.

Top of Page

Control adjustment

Westbound

The eastbound estimates for indirect trade were calculated by subtracting Canada's indirect import figures from the respective Chinese indirect export figures. To some degree this limited the problem of over-estimations of the adjusted figures due to double counting. This methodology was not possible for the indirect trade adjustments in the westbound direction. While it was possible to distinguish between Chinese direct and indirect imports, this was not possible to distinguish between direct and indirect exports using the Canadian data. Canada only reports one country in its export figures, the country of last known destination. Consequently, all of China 's reported indirect trade values for the respective years were included in the indirect trade adjustments. Double counting was inevitable.

A possible solution to limit the degree of double counting is to control for the occurrences where Country A's exports plus Country B's indirect imports are greater than Country B's imports.

In an ideal world, Country A's exports should equal country B's imports. However, this is generally not the case, due to indirect trade and other factors. It is often the case that the exporter does not know where the final destination of the goods will be. Consequently, the exporter indicates the intermediary country as the country of last known destination. Imports are, in general, more accurately reported than exports because of the requirements for tariff assessment and the application of trade agreements. Thus, the importing country will have a record of the country of origin and of the intermediary countries involved in the trade and these records will be reported more accurately than those of exports.

If we accept the argument that one of the primary causes of differences found between partner country trade figures is the misallocated reporting of export goods then the second best situation would be where Country A's exports plus Country B's indirect imports equal Country B's total imports. For example, Canada exports $10 worth of goods and China reports indirect imports of $5 and total imports of $15. It would also be acceptable where Canada exports $10 worth of goods and China reports indirect imports of $2 and total imports of $15, that is, where the sum of exports plus indirect exports is less than total imports. This would imply that while misallocated trade is a cause for the discrepancy there are still some other factors to take into consideration. However, if the sum of Canada's exports plus China's indirect imports is greater than China's total imports, (for example, if Canada's exports are $10, China's indirect imports are reported as $7 for a total of $17 and China's total imports are reported as $15), then this would indicate that some double counting is involved (i.e. $2); Canada has captured some of the intermediary trade in its statistics but not all of it.

An attempt was made to control for double counting in the westbound trade using this premise. The following steps were taken:

1. The Chinese reported import values were adjusted for insurance and freight by applying the estimated insurance and freight rates, at the two digit HS level, to the corresponding commodities;

2. For adjusted Chinese imports greater than Canadian domestic exports (CN^adjM > CA^DX) the adjusted Chinese reported indirect imports were added to the Canadian reported domestic exports (CN^adjIndM + CA^DX) and compared to the adjusted Chinese total imports (CN^M);

3. Residuals (R) were calculated for CN^adjIndM + CA^DX > CN^M, where R equals CN^adjM - (CN^adjIndM + CA^DX);

4. The residuals were added together to determine a 'control' value.

Top of Page

Price mark-up adjustment

Hong Kong mark-up

Eastbound

Estimates for the Hong Kong mark-up values were made using data provided by the Hong Kong Census and Statistics Department. For re-export purposes, the department collects data concerning the two countries involved in the re-exports of goods through its economy, namely country of origin and country of destination. This proved useful when trying to determine price mark-up estimates.

To determine an estimate for Hong Kong mark-up in prices for eastbound trade, Hong Kong 's re-export data for goods of Chinese origin re-exported to Canada and Hong Kong 's reported imports from China were used. The Hong Kong mark-up ratios were calculated as follows:

where at the HS 6 level:

HK^RXCA UV: the unit value of the Hong Kong re-exports of goods of Chinese origin that were exported, via Hong Kong to Canada;

HK^RXCA Qty: the quantity of the Hong Kong re-exports of goods of Chinese origin that were exported, via Hong Kong to Canada;

HK^MCN UV: the unit value of Hong Kong imports of Chinese origin;

UV: the unit value of the good, defined as value divided by quantity.

The commodities of the two data sets were matched at the HS 6 level. If a match was not found that commodity was not included in the calculation used to determine the mark-up ratio.

The commodities whose quantity and/or value were zero or whose unit of measure (UOM) was zero were not included in the calculations. According to Hong Kong data descriptions, a UOM of zero indicates a mixed grouping of goods, consequently a unit value for the grouping cannot be determined.

Unit values were calculated for the remaining commodities and the ratio was then calculated. The Hong Kong mark-up ratio for 2002 and 2003 were 1.43 and 1.40 respectively.

Westbound

Data from a Hong Kong survey was used to determine mark-up estimates for the westbound trade. This mark-up ratio was applied to the value of the Canadian published imports from China via Hong Kong to determine the adjusted values for Canadian imports via Hong Kong.

Adjusted Canadian imports via Hong Kong :

Top of Page

Re-export adjustment

Eastbound

Chinese export statistics are based on country of ultimate destination and include domestic goods, i.e., goods of Chinese origin, and re-exports, i.e., goods of foreign origin that have entered Chinese consumption and have subsequently been sold without any substantial transformation occurring in China. Canadian import data is based on the country of origin principle.

The Chinese export data does not distinguish between domestic exports and re-exports. However, Canadian data documents both the country of origin (COO) and the country of export (COE) i.e., the country from which the goods are shipped directly to Canada. Consequently it is possible to estimate Chinese re-exports using Canadian data. The values for transactions that showed China as country of export (COE equals CN) but not country of origin (COO does not equal CN) were totalled and used to estimate Chinese re-exports. For the purpose of this reconciliation study, the estimated Chinese re-exports were added to the Canadian published imports.

Westbound

Canadian export data distinguishes domestic exports and re-exports. Therefore, Canadian export data could be used to adjust the westbound flow by adding these amounts to Chinese published imports.

Top of Page

Valuation adjustment

Canada reports its exports F.O.B. (Free on Board) at the Canadian border. This method excludes insurance and freight costs from point of exit. China reports imports on a C.I.F. (Cost, Insurance, and Freight) basis, which does include all insurance and freight cost to the Chinese port of destination. Estimates of insurance and freight costs are necessary to align the respective data. While China reports imports on a C.I.F. basis it does not capture the freight and insurance costs separately. For the purpose of this reconciliation study, data from the United States Census Bureau (USCB) were used to estimate the insurance and freight costs.

The USCB records import values on both a C.I.F. and F.O.B. basis. It also collects separate records for freight and insurance. The freight and insurance ratios for each commodity at the HS2 level were estimated using the following:

Insurance and Freight ratio:

The resulting ratio was then applied to the Chinese imports (C.I.F. values) to determine the F.O.B. values of Chinese imported goods at the HS2 level as follows:

Adjusted Chinese Imports equals Chinese Imports multiplied by (one minus Insurance and Freight Ratio).

There are a couple of considerations to keep in mind:

1. The insurance and freight costs apply to trade flows which are in the reverse direction from those of the Chinese import data that is; eastbound (CN to US) trade versus westbound trade (US to CN). The assumption is made that eastbound and westbound insurance and freight costs are similar.

2. The commodity trade is not necessarily the same. China does not necessarily import the same commodities from Canada that it does from the U.S. Similarly, the U.S. and Canada do not necessarily export the same type of commodities to China.

Top of Page

Residual adjustment

The residual is calculated by summing all reconciliation adjustments and subtracting the total from the published import figure. It therefore largely comprises two main aspects: errors or deficiencies in the other reconciliation adjustments and export undercoverage.