Reconciliation of the U.S.-Canadian current account, 1992-93.

DiLullo, Anthony J. ; Laliberte, Lucie

THE RECONCILIATION of the U.S.-Canadian current account for 1992 and 1993 resulted in a smaller U.S. surplus, or Canadian deficit, for 1992 and a shift to a small U.S. deficit, or Canadian surplus, for 1993. Before reconciliation, the U.S. current-account surplus with Canada is $3.3 billion for 1992 and $1.4 billion for 1993; the corresponding Canadian estimates are $1.9 billion for 1992 and $1.7 billion for 1993. After reconciliation, the U.S. surplus for 1992 is reduced to $1.0 billion, and the U.S. surplus for 1993 changes to a small deficit (chart 1, table 1).

[TABULAR DATA 1 OMITTED]

This article presents details of the reconciliation of the U.S.-Canadian current account for 1992 and 1993 by Statistics Canada and the Bureau of Economic Analysis (BEA).(1) The reconciliation is undertaken because of the extensive economic links between Canada and the United States and the need to explain differences in the Canadian and U.S. published estimates of bilateral current-account transactions. In principle, the bilateral current account of one country should mirror the bilateral current account of the other country.

The reconciliation process has improved the accuracy of the published estimates of transactions between Canada and the United States and has increased the efficiency of producing the estimates. The improvements have been accomplished through the exchange of data between the two countries and the development of improved estimating techniques. The exchange of data covers over 8o percent of the value of Canadian and U.S. current-account transactions. This year, a change in the definition of direct investment income in the Canadian published estimates eliminated one of the remaining major definitional differences between the U.S. and Canadian published estimates.

After reconciliation, revisions are incorporated into the Canadian and U.S. published estimates as far as possible. A complete exchange of data or substitution of reconciled estimates for published estimates is not feasible, because of definitional and methodological differences and because estimates of transactions with third countries would be affected in some cases. In addition, protecting the confidentiality of source data bars the exchange of data for some transactions.

The rest of this article consists of two parts. The first part briefly discusses the reconciled balances for 1992-93. The second part summarizes the methodology for reconciliation and discusses the major reconciliation adjustments to the U.S. and Canadian current accounts.

Reconciled Balances for 1992-93

For 1992, the reconciled U.S.-Canadian current-account balance is a U.S. surplus, or Canadian deficit, of $1.0 billion; the U.S. published estimate is a surplus Of $3.3 billion, and the Canadian published estimate is a deficit of $1.9 billion. For 1993, the balance is not fully reconciled. The provisional U.S. estimate of the reconciled balance is a U.S. deficit, or Canadian surplus, of $0.3 billion; the provisional Canadian estimate is a U.S. deficit, or Canadian surplus, of $0.1 billion. By comparison, the U.S. published estimate for 1993 is a $1.4 billion surplus, and the Canadian published estimate is a $1.7 billion deficit. The difference between the provisional reconciled estimates reflects the preliminary nature of the 1993 data, as well as difficulties in reconciling some service and direct investment estimates related to insurance transactions. In the past, those differences have been offsetting; in 1993, they are not.

Adjustments to the U.S. southbound estimates are larger than those to the northbound estimates for both years (tables 2 and 3).(2) The increases in U.S. payments primarily reflect definitional adjustments to merchandise trade, such as the addition of Canadian reexports to U.S. merchandise imports, and statistical adjustments to transportation (inland freight) and "other services" for undercoverage. The largest adjustments to U.S. estimates of receipts, such as the reclassification of inland freight and statistical adjustments to transportation and investment income, are partly offsetting. For 1992, the reconciliation adjustments resulted in a reduction of $2.3 billion in the U.S. published surplus; for 1993, they resulted in a shift from a $1.4 billion surplus to a small deficit (tables 4 and 5).

[TABULAR DATA 2-5 OMITTED]

The only large overall adjustments to the Canadian estimates were to the 1993 northbound estimates. The reconciled balances mainly reflect a reduction in the Canadian published deficits. For 1992, the reconciliation adjustments resulted in a reduction of $0.9 billion in the Canadian published deficit; for 1993, they resulted in a shift from a $1.7 billion deficit to a small surplus.

Reconciliation Adjustments

Reconciliation adjustments to each country's published estimates are classified in three major categories--definitional, methodological, and statistical--which reflect the differences that occur in the published estimates.(3)

Definitional and methodological adjustments are required because of differences in the definitions and methodologies used to compile the international accounts in Canada and the United States. These differences primarily reflect each country's requirements to integrate the external accounts with the domestic accounts and differences in compilation conventions due to institutional factors. To achieve reconciliation, a common definition is selected, and methodologies are adjusted to a common basis. The choice of one definition or methodology over another does not necessarily indicate agreement on the correct definition or the most appropriate methodology. Often, the choice is based on practical considerations, such as the availability of data.

Statistical adjustments reflect differences in data sources and estimation techniques. There are four types of statistical adjustments. First, some adjustments are based on an evaluation of the quality and coverage of the source data. When one country's data are believed to be of better quality, the better data are used to develop the reconciliation adjustments. Second, some adjustments are based on detailed information that is available from one country but not the other. Third, adjustments are made to reflect revisions in source data that become available after publication of the estimates. Fourth, some adjustments are essentially pragmatic, particularly when it is not possible to clearly establish the merits of one country's data relative to the other country's data. In such cases, reconciled values are developed that are believed to be within a reasonable range of measurement error.

The following sections present a discussion of the major reconciliation adjustments made to the current account. Although numerous adjustments are made, only those that involve important definitional, methodological, or statistical differences are discussed. Definitional and statistical adjustments affect the current-account balance because they affect the value of transactions. Methodological adjustments, such as the reclassification or the grossing or netting of transactions, are necessary to achieve common treatment, but because they are offsetting, they do not affect the current-account balance.

Merchandise trade

Most of the differences between Canadian and U.S. published estimates of merchandise trade stem from different treatment of the source data (tables 6 and 7).(4) Among the definitional adjustments, U.S. imports are redefined to include Canadian reexports in order to align them with Canadian estimates of exports to the United States. U.S. published estimates, which are on a country-of-origin basis, attribute Canadian reexports to third countries rather than to Canada; the Canadian published estimates are on a country-of-shipment basis and thus do not require a similar adjustment. Definitional adjustments are made to the Canadian published estimates to eliminate timing differences (including progress payments on certain military equipment imported from the United States) and valuation differences.

[TABULAR DATA 6 & 7 OMITTED]

Among the methodological adjustments, inland freight charges (freight charges on overland shipments of exports and imports from the plant to the border) are reclassified from merchandise trade to the transportation accounts. This adjustment is made in order to value merchandise trade in the U.S. accounts at the plant, rather than at the border as in the U.S. published estimates. In the Canadian published estimates, merchandise trade, with the exception of natural gas exports, is valued at the plant, and inland freight charges, except for those on natural gas exports, are included in transportation; for reconciliation, inland freight charges on Canadian exports of natural gas are reclassified to transportation in the Canadian accounts.

Other methodological adjustments include the reclassification of equipment repairs from services to merchandise trade in the U.S. accounts. In the U.S. published accounts, all equipment repairs are classified in services, though some are initially reported in the merchandise trade source data.

Services

Travel and passenger fares.--The Canadian and U.S. published estimates of travel and passenger fares are based mostly on the same source data and on common definitions. Thus, the reconciliation adjustments are statistical. The U.S. estimates are adjusted to exclude some transactions for cruise travel that may duplicate similar transactions in the Canadian source data and to account for differences in the timing of publication of revised estimates by Statistics Canada and BEA.

Transportation.--The adjustments to transportation are largely methodological (reclassification) and statistical (tables 8 and 9). The methodological adjustments include the previously mentioned reclassification of inland freight charges from merchandise trade to transportation. In addition, expenditures for port services by air and rail carriers are reclassified from business services to transportation in the Canadian published accounts to align them with the U.S. treatment. Statistical adjustments are made to reduce the U.S. northbound estimates of inland freight to the Canadian level because the Canadian published estimates are believed to be more accurate. Adjustments are made to the U.S. southbound estimates of inland freight to add Canadian estimates of inland freight on U.S. natural gas and newsprint imports, which are not fully covered in the U.S. source data.

[TABULAR DATA 8 & 9 OMITTED]

A single definitional adjustment is made to the Canadian southbound estimates to delete estimates of freight charges on U.S. exports in transit in Canada. In the U.S. treatment, those charges are viewed as payable by the importer and thus as transactions between Canada and third countries.

"Other services."--"Other services" include transactions in royalties and license fees and in a variety of other services between affiliated, or related, persons and between unaffiliated persons; they also include government transactions (tables 10 and 11). Affiliated services are reconciled at a highly aggregated level because of the lack of detailed information by type of transaction in the U.S. source data. By contrast, the source data on unaffiliated transactions contain a large amount of detail by type of transaction, so a detailed reconciliation of unaffiliated transactions is undertaken.

[TABULAR DATA 10 & 11 OMITTED]

Definitional adjustments are made to the Canadian estimates of affiliated and unaffiliated transactions to exclude nonresident taxes. In the U.S. accounts, nonresident taxes are included on a global basis, but they are not allocated by country. In addition, commissions on merchandise trade are removed from the Canadian estimates to align them with the U.S. treatment.

Methodological adjustments are made only to unaffiliated transactions. They include the reclassification of air and rail carriers' port services to transportation in the Canadian accounts. In the U.S. northbound accounts, equipment repairs are reclassified to merchandise trade, and medical services covered by Canadian Provincial health insurance programs are reclassified to Canadian government services. Other methodological adjustments include netting the U.S. estimates of northbound communications transactions against southbound transactions. detailed comparison of the Canadian and U.S. estimates of communication transactions cannot be undertaken because of the confidentiality of source data. The Canadian estimates of insurance transactions are adjusted to a net basis for comparability with the U.S. estimates; however, insurance transactions cannot be fully reconciled because of differences in accounting conventions and data collection methods for the insurance industry in Canada and the United States.

Statistical adjustments to the U.S. estimates include increases to account for undercoverage of both affiliated and unaffiliated transactions. Statistical adjustments to the Canadian estimates are made mostly to affiliated transactions, which are reduced to reflect removal of an estimate for undercoverage and to eliminate possible duplication of tooling charges in merchandise trade and services.

Government transactions require only statistical adjustments, which reflect timing differences in the publication of revised estimates. Estimates of government transactions are exchanged between Statistics Canada and BEA.

Investment income

Direct investment income.--This year, the reconciliation of direct investment income includes the total income--distributed earnings, reinvested earnings, and interest--of incorporated and unincorporated affiliates. In earlier years, the reconciliation excluded reinvested earnings of incorporated affiliates because they were not included in the Canadian published estimates of direct investment income. The inclusion of reinvested earnings in the Canadian published estimates eliminates one of the major definitional differences in the Canadian and U.S. published estimates of investment income.

Statistical adjustments were the largest type of adjustment made to the direct investment income estimates (tables 12 and 13). U.S. estimates of northbound reinvested earnings of incorporated affiliates reflect higher earnings, or smaller losses, than the Canadian estimates. Until further research on the reasons for the differences can be undertaken, the estimates were reconciled at an arbitrary value midway between the U.S. and Canadian published estimates. For southbound reinvested earnings, U.S. estimates are used for reconciliation. For distributed earnings (dividends), Canadian estimates are used to reconcile northbound and southbound estimates. Statistical adjustments to earnings of unincorporated affiliates, which are reconciled in aggregate, are made mainly to earnings in the banking, insurance, and real estate industries for both northbound and southbound estimates.

[TABULAR DATA 12 & 13 OMITTED]

Definitional adjustments are made to the Canadian published estimates only. The Canadian estimates are adjusted to exclude nonresident taxes; in addition, the Canadian estimates of earnings of insurance affiliates are adjusted to an accrual basis from a cash basis.

Methodological adjustments are made to the Canadian estimates; these include the reclassification of certain dividends from portfolio income to direct investment income and the adjustment of interest receipts and payments to a net basis for comparability with the U.S. estimates.

"Other investment (portfolio) income."--The reconciliation adjustments to "other investment income" are mostly statistical to account for differences in source data (tables 14 and 15). Definitional adjustments consist primarily of the removal of nonresident taxes from the Canadian estimates.

[TABULAR DATA 14 & 15 OMITTED]

Methodological adjustments are made primarily to the estimates of income on bank claims and liabilities. With one exception, the Canadian estimates are converted from a net basis (receipts on claims less payments on liabilities) to a gross basis to align them with the U.S. treatment; the exception is income receipts and payments between affiliated U.S. and Canadian banks, which are reconciled on a net basis. On a gross basis, the Canadian estimates substantially exceed the U.S. estimates, but on a net basis, they are almost identical. The divergence in the estimates on a gross basis may be due to differences in reporting definitions for banks in Canada and in the United States.

Several statistical adjustments are made, primarily to income on securities. First, the U.S. estimates of northbound dividends are raised to the Canadian level, and the Canadian estimates of southbound dividends are raised to the U.S. level; these adjustments reflect the general assumption that a country's source data on payments of dividends are more comprehensive than its source data on receipts. Second, U.S. northbound estimates of income receipts on holdings of Canadian bonds are raised to the Canadian level; the Canadian estimates are based on an inventory of individual bonds held by nonresidents that is more detailed and more comprehensive than the U.S. inventory. Third, the Canadian and U.S. estimates of payments on U.S. Government liabilities are arbitrarily adjusted to a midpoint. The Canadian estimates are lower than the U.S. estimates because compilers of the Canadian accounts assume that some U.S. Government securities are purchased by Canadian parents on behalf of their U.S. insurance affiliates; thus, the income would accrue to U.S. residents. Based on indications in the U.S. source data, compilers of the U.S. accounts assume that those purchases and the income are entirely for the accounts of the Canadian parents.(5) Finally, some adjustments are made to the Canadian and U.S. estimates to reconcile miscellaneous commercial transactions.

Unilateral transfers

In a definitional adjustment, the Canadian estimates of unilateral transfers are reduced by the removal of nonresident taxes. In a methodological adjustment, the U.S. estimates are converted to a gross basis from a net basis to align them with the Canadian treatment. Small statistical adjustments are made to the U.S. northbound estimates and to the Canadian southbound estimates to compensate for a lack of coverage in the source data.

(1.) The reconciliation of the current account has been undertaken each year since 1970. Summary results of the reconciliations were published in the United States in the following issues of the Survey of Current Business: June 1975, September 1976 and 1977, December 1979, June 1981, December 1981 through 1991, November 1992, and October 1993. In Canada, the results were published in the following issues of Canada's Balance of International Payments (catalogue 67-001), a publication of Statistics Canada: Fourth Quarter 1973, Second Quarter 1976 and 1977, Third Quarter 1978 and 1979, First Quarter 1981, and Third Quarter 1981 through 1993.

(2.) In this article, the term "northbound" refers to U.S. receipts, or Canadian payments; "southbound" refers to Canadian receipts, or U.S. payments. All values are expressed in U.S. dollars.

(3.) A detailed article on the methodology used to reconcile the U.S.-Canadian current account was published by BEA in the November 1992 issue of the Survey and by Statistics Canada in Reconciliation of the Canada-United States Current Account, 1990-91. Statistics Canada also published a shortened version in the December 1992 Canadian Economic Observer and in Canada's Balance of International Payments, Third Quarter 1992

(4.) The source data are the same for both countries, except for those used by Statistics Canada to compile petroleum and natural gas exports. The data, except as noted, are compiled from U.S. and Canadian customs documents filed by U.S. and Canadian importers. U.S. merchandise imports are compiled from U.S. customs documents, and U.S. exports (Canadian imports) are compiled from data provided from Canadian customs documents. Similarly, Canadian merchandise imports are compiled from Canadian import documents, and Canadian exports (U.S. imports) are compiled from data provided from U.S. customs documents.

(5.) In 1993, the difference between the U.S. and Canadian estimates of payments on U.S. Government liabilities widened considerably. The U.S. estimate increased substantially, whereas the Canadian estimate decreased. The U.S. estimate reflects a much larger increase in Canadian purchases of U.S. Government securities than the Canadian estimate, and a much smaller decline in yields.