Activities of Canadian Multinational Enterprises Abroad
Detailed information for 2018
This survey measures the sales of goods and services, employment levels and the assets and liabilities of Canadian multinational enterprises abroad.
Data release - December 2, 2020
- Questionnaire(s) and reporting guide(s)
- Data sources and methodology
- Data accuracy
This survey measures the sales of goods and services, employment levels and the assets and liabilities of Canadian multinational enterprises abroad. Throughout this document, the terms "Canadian majority-owned affiliates abroad," "(Canadian) majority-owned foreign affiliates (abroad)" and "MOFAs" refer interchangeably to Canadian multinational enterprises abroad.
Selling goods or services through MOFAs is a means for Canadian companies to market their products internationally. In the case of goods, the products sold by majority-owned foreign affiliates may be produced in Canada or produced abroad.
Reference period: Calendar year
Collection period: The collection period is generally from April to November of the year after the reference period.
- Economic accounts
Data sources and methodology
The universe for this survey is comprised of foreign affiliates abroad that are majority-owned (i.e. more than 50% of the voting shares) by a business that resides in Canada.
In order to be consistent with the international practice for measuring foreign affiliate trade statistics, only the data for majority-owned foreign affiliates (MOFAs) are included. Sales and employment figures of MOFAs are fully attributed; there is no adjustment for less than 100% ownership.
The questionnaire BP-CIA (Canadian Investments Abroad), which is used to collect data on Foreign Direct Investment Abroad, incorporates specific questions for compiling the data on Canada's majority-owned foreign affiliate statistics.
The survey was last redesigned in 2011 and went through extensive testing through Questionnaire Design and Research Centre. Field testing with respondents was also conducted.
The BP-CIA questionnaire (Canadian Investment Abroad), which contains questions pertaining to majority-owned foreign affiliates (MOFAs) as well as other questions about Canadian investment abroad, is sent to Canadian enterprises known to have or believed to have a direct investment relationship (i.e. 10% or more of the voting equity or equivalent) in one or more non-resident corporations, partnerships or joint ventures. For the statistics on MOFAs, only the majority-owned foreign affiliates are included. Foreign affiliates that are not majority-owned, but are at least 10% owned by a Canadian enterprise, are included in Canadian direct investment abroad positions nonetheless. See record number 1537 for more information about Canada's international investment position.
For reference year 2018 around 900 firms received the BP-CIA questionnaire (Canadian Investment Abroad). This survey is annual and is believed to cover close to 100% of the target population.
The BP-CIA (Canadian investment abroad) survey uses a stratified design with simple random sample selection in each stratum. The first stratification was the must-take portion known as the units selected with certainty. For the non must-take portion, the stratification was done by industrial groups and geographical regions, and the population was further stratified by size (take all, take-some 1 and take-some 2).
For the non must-take portion, the sample was allocated to the two take-some groups, by industrial group and geographical region. Also, a minimum number of four enterprises for each take-some stratum was set. The sample selection in the take-some strata was random. The sample size fluctuates from cycle to cycle; for 2018 reference year the sample size was approximately 900 units.
Responding to this survey is mandatory.
Data are collected directly from survey respondents.
Data are collected from both paper and electronic surveys. Response is mandatory for the surveyed entities. Respondents are instructed to complete their paper or electronic surveys within four weeks of receiving the questionnaire or the email invitation for completing the electronic questionnaire.
A follow-up is done after an additional two weeks if no response is received, followed by phone calls.
Data are captured into a system by an internal centralized department. Electronic survey data is automatically captured into systems.
Both official languages are available to respondents.
The following administrative data sources from the Canada Revenue Agency are used in the sampling process:
T2 schedule 9 - Related and Associated Corporations
T106 - Information Return of Non-Arm's Length Transactions with Non-Residents
Other sources - A variety of other sources are used such as financial press, business publications, company reports, etc.
View the Questionnaire(s) and reporting guide(s) .
The data received from questionnaire respondents are submitted to edits during data collection, data capture, and data processing to ensure consistency and coherence.
Many types of edits are performed at the micro level:
-edits built-in to the electronic questionnaire to flag extreme values, missing values, as well as consistency edits between questionnaire sections;
-edits in the capture system for mandatory fields (primarily applies to paper questionnaire);
-edits in the processing system to flag period over period consistency issues, and non-mandatory fields.
When necessary, there is follow up with the respondent to verify or correct the data.
Macro edits are also performed:
-Further time-series analysis at the macro leads to analysis at the micro level, which may lead to micro edits and follow-up.
-Administrative and external data sources are also used at the macro level for data confrontation to provide additional error detection.
The imputation process examines the relationship between foreign direct investment and Canadian majority-owned foreign affiliate variables such as sales and employment to impute the missing values. If all the variables for a given record contain missing information, then that unit is considered to be a non-respondent. In this case a re-weighting takes place in order to account for this unit in the estimates.
For some non-responding units that are deemed to be significant in terms of their overall contribution to the estimates (based on either previously reported data or publicly available information), data is researched on a case by case basis and estimated largely from information contained in their annual reports, financial statements or websites.
For direct investment, population parameters are defined by historical survey data together with tax data that indicates international activity. This is done at the enterprise level, which is linked to the central Business Register.
The reported values for sampled units are multiplied by a sampling weight in order to provide an estimate for the entire population. The sampling weight is calculated based on certain factors, such as the probability for a unit to be selected in the sample and that responding units will have to represent sampled units that did not respond.
Before the Canadian majority-owned foreign affiliate data are published, several steps are taken to ensure the quality of the estimates.
The final estimates are compared to the pre-imputation data. Any significant differences are investigated, and if necessary adjustments are made when the imputation or estimation process results are judged to be unacceptable.
Counterpart data submitted to supranational organizations by national statistical agencies are also confronted, along with other sources such as the financial press, business publications and company reports.
Statistics Canada is prohibited by law from releasing any information it collects that could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.
Revisions and seasonal adjustment
At the time of publication, data for the two preceding years are subject to revision. The data are not normally revised again except when comprehensive revisions are carried out.
Seasonal adjustment is not applicable to annual data series.
Non-sampling errors - While considerable effort is made to ensure high standards throughout all stages of collection and processing, the resulting estimates are inevitably subject to a certain degree of non-sampling error. Examples of non-sampling error are coverage error, data response error, non-response error and processing error.
Coverage error can result from incomplete listing and inadequate coverage of the population of Canadian transactors that engage in international direct investment. Coverage from tax data is comprehensive and greatly limits coverage error.
Data response error may be due to questionnaire design, the characteristics of a question, inability or unwillingness of the respondent to provide correct information, misinterpretation of the questions or definitional problems.
Non-response error is related to respondents that may refuse to answer, are unable to respond or are too late in reporting. In these cases, data are imputed. The extent of any imputation error decreases with increases in the response rate and attempts are therefore made to obtain as high a response rate as possible. The response rate for the BP-CIA questionnaire, which is used to collect foreign affiliate trade statistics, is around 70%. Analysts keep in contact with the respondents to try to maximize the response rate. Slow reporting is often an issue.
Processing error may occur at various stages of processing such as data entry, editing and tabulation. Measures have been taken to minimize these errors. Data entry and edit are performed simultaneously due to the questionnaire design which allows errors to be quickly seen. Historical ratios also aid in eliminating outliers created by data entry. Tabulation is automated to eliminate human error.
Direct investment surveys have response rates of approximately 70%-75%.
While considerable effort is made to ensure high standards throughout all stages of collection and processing, the resulting estimates are inevitably subject to a certain degree of non-sampling error. Examples of non-sampling error are coverage error, data response error, non-response error and processing error.
Total non-response is defined as units that did not answer the questionnaire. In order to limit total non-response, total non-response units were examined and every effort was made to impute values for them.
- User Guide: Canadian System of Macroeconomic Accounts
This guide provides a detailed explanation of the structure, concepts and history of the Canadian System of Macroeconomic Accounts.
Last review: June 22, 2018