Capital and Repair Expenditures, Actual, Preliminary Actual and Intentions (CAPEX)

Detailed information for 2007

This survey collects data on capital and repair* expenditures in Canada. The information is used by Federal and Provincial governments and agencies, trade associations, universities and international organizations for policy development and as a measure of regional activity. *Organizations reporting in the preliminary actual, intentions were not asked for repair expenditures.

Data release - February 28, 2007 (intentions); February 27 2008 (preliminary actual); February 25, 2009 (actual)

• Questionnaire(s) and reporting guide(s)
• Description
• Data sources and methodology
• Documentation

Description

This annual survey collects data on the intentions for capital investment and the expenditures for the previous two years; on occasion, where economic changes justify the need, data on the revised intentions are also collected for the current year. Information on capital spending provides a useful indication of market conditions both in the economy at large and in particular industries. Since such expenditures account for a large and relatively variable proportion of gross domestic expenditures, the size and content of the investment program provides significant information about demands that have been placed upon the productive capabilities of the economy during the period covered by the survey. In addition, information on the relative size of the capital expenditures program planned, both in total and for individual industries, gives an indication of the views management hold on the future market demands in relation to present productive capacity. Inputs are used by the Canadian System of National Accounts, particularly in the Gross Domestic Product (GDP) and Balance of Payments. The information is used by Federal and Provincial governments and agencies, trade associations, universities and international organizations for policy development and as a measure of regional activity.

Reference period: Fiscal year

Subjects

• Business, consumer and property services
• Construction
• Economic accounts
• Machinery and equipment
• Non-residential building construction
• Non-residential engineering construction
• Repair and maintenance

Data sources and methodology

Target population

The target population comprises all business and government entities operating in Canada. Outlays for used Canadian assets are excluded since they constitute a transfer of assets within Canada and have no effect on the aggregates of our domestic inventory. Assets imported from outside Canada are included as they increase our domestic inventory.

Instrument design

The questionnaire was redesigned in 1985 and has remained largely unchanged since then.

Sampling

This is a sample survey with a cross-sectional design.

The target population consists of all Canadian businesses in Canada. The frame is the list of establishments from which the portion eligible for sampling is determined and the sample is taken. The frame is maintained by Statistics Canada's Business Register and is updated using administrative data.

Prior to the selection of a random sample, establishments are classified into homogeneous groups (i.e., groups with the same NAICS codes, same geography (province/territory). Quality requirements are targeted, and then each group is divided into sub-groups called strata: take-all and take-some.

In the take-all strata, all units are sampled with certainty whereas the take-some strata is sampled using the simple random process and a minimal fraction of 1% is imposed and a minimum of 3 units is required where possible. The sample size for the survey is 27,000 units

Data sources

Responding to this survey is mandatory.

Data are collected directly from survey respondents, extracted from administrative files and derived from other Statistics Canada surveys and/or other sources.

Two survey periods are organized and timed to collect three sets of annual data related to investment intentions, preliminary actual expenditures and actual capital and repair expenditures. The investment intentions for year Y and the preliminary actual expenditures for Y-1 are collected, using a combined questionnaire, from October of Y-1 to January of Y; the actual capital and repair expenditures for Y-2 are collected from March of Y-1 to October of Y-1. Establishments selected for the sample receive either the regular survey questionnaire (short or long form), a specialized survey questionnaire (short or long form) or the new project questionnaire. The type of questionnaire an establishment receives depends on the industry, the expected level of expenditure, the set of data being collected and whether or not the establishment is classified as a new project. Respondents are asked to report expenditures for their 12 months fiscal period for which the final day occurs between April 1 of the reference year and March 31 of the following year.

Once the sample has been selected, questionnaires are mailed out. Units that have not responded are subject to mail, fax and telephone follow-up to ensure the data is obtained. Low response rate to the survey within a specific industry and province/territory represents the primary reason for follow-up. Initially, a general reminder is sent in the form of a mailout to the entire delinquent portion of the sample. If non-response continues, establishments in areas of lowest coverage are solicited by telephone for the return of the completed questionnaire.

In addition to the data collected by the above mentioned questionnaires, data which are incorporated into the production of capital expenditure estimates are also obtained from other surveys conducted by Statistics Canada and from provincial government administrative files. As well, some estimates of capital expenditures for some industries are produced based on indicators of production, consumption and costs associated with operation in those industries.

View the Questionnaire(s) and reporting guide(s).

Error detection

After the questionnaires have been completed and returned, the process of quality assurance continues through data editing. Data are screened at the micro level for internal survey over survey and year over year inconsistencies. On the actual questionnaire for respondents receiving long forms, asset detail edits identify details which are inconsistent with previous questionnaire returns or with assets commonly used in the respondent's industry. In addition, the outlier detection module is run before and after imputation. After imputation, this is done with the imputed data and permits detection of outliers among imputed data. The Hidiroglou-Berthelot (1986) method is used to detect outliers.

Imputation

Non-respondents are imputed where possible using their previous response for the same year which is adjusted by a factor for survey over survey rate of change.

Estimation

The ratio estimator is used for estimation with revenue being the auxiliary variable. This method ensures that the final weight multiplied by the income for each unit in the sample matches the known total for the income variable for the entire population in the group. Some adjustments are made for non-response. Variance is estimated using Taylor's linearization formula in the case of ratio estimator.

The final weight of the 2009 Capital Expenditure Intentions Survey is used as the starting point for estimation for the second phase of the 2009 survey. Each sample unit was selected based on a known probability. The inverse of that probability is the plan weight. To counter nonresponse effects, a second adjustment is done. This is equal to the inverse of the unweighted response rate observed at the strata level. A calibration factor is then calculated for each stratum. The total amount of capital expenditure intentions published for 2009 is used as a control total for each survey stratum. The final weight for each respondent is the product of its final weight resulting from the 2009 Capital Expenditure Intentions Survey, its plan weight, its nonresponse adjustment and its calibration factor. This final weight is used to produce estimates and their associated coefficients of variation. The Horvitz-Thompson estimator is used for two-phase sample plans. The SAS StatMx macros set produces the estimates and coefficients of variation for the variables being studied. Estimation is done at the strata level only.

Quality evaluation

The estimation process produces a comprehensive data set for the surveyed and non-surveyed portions of the universe (frame). After the estimation process, trend analysis for the various industries can be done. Commencing with an evaluation of the year over year (or percentage) change in each industry, provinces/territories that have industries or sub-industries experiencing unusual activity are highlighted. In addition, this type of analysis also identifies industries which have the largest impact on Canadian aggregates.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects which 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

Up to reference period 2006, provided data was adjusted to represent the calendar year. The calendarization method was used to reduce response burden, so the respondent could provide data on a fiscal year basis. Once collected and edited, the data was calendarized to generate data for the January to December period for the reference year for cases where respondent reported data on another period. To prevent the production estimates linked to many fiscal periods, calendarization was done. The Cholette (1984) method was used to break the data in monthly portions and extrapolate.

Starting with reference year 2007, the data published is reflecting data provided by respondents based on their fiscal period and no calendar adjustment is applied. For further information on the impact of this methodological change on the time series, an analytical paper is available on request.

Documentation

• Sources and Methods - Capital and Repair Expenditures, Actual Preliminary and Intentions
  This documentation is valid from 2001 until further notice.
  • Format: Sources and Methods - Capital and Repair Expenditures, Actual Preliminary and Intentions - ARCHIVED - PDF, 240.13[ARCHIVED - PDF, 240.13 kb]