What is the purpose of the Quarterly Industry Revenue Indices?
What data are produced by QIRI?
How does QIRI determine which industries to survey?
Why are data for only some industries available?
Will more industries be available in future?
What is the North American Industrial Classification System (NAICS)?
What is an index?
Why are the data indexed?
How are quarterly indices annualized?
How reliable are the data?
What are the data sources for QIRI?
What is the difference between simple and complex units?
Why are the data adjusted for seasonality?
What is the purpose of the Quarterly Industry Revenue Indices?
The Quarterly Industry Revenue Indices (QIRI) provide quarterly indexed rates of change in operating revenues for selected industries in business and consumer services in order to track their quarterly economic performance. As there are few sub-annual output measures for the services industries, the QIRI fills this gap for policy makers requiring more timely information to better assess current economic conditions. The need for such an indicator arises from the fact that services-producing industries now account for approximately 70% of Canadian gross domestic product (GDP).
In a paper prepared for Statistics Canada's 2005 Economic Conference, a rationale for sub-annual indicators of service industries was provided based on an examination of pertinent research and reviewing existing sub-annual data sources at Statistics Canada (Fiori, McKeown & Taktek, 2005).1 Despite its dominance in developed economies, the services sector has been less analyzed than the goods-producing sector due to data and measurement limitations. Statistics Canada has relied heavily on employment statistics to measure the output of the services sector, particularly at the sub-annual level. Estimates of monthly GDP for industries representing more than one-third of the economy use employment (a lagging indicator) either exclusively or partly as an indicator for lack of better data. Excluding distributive services (NAICS 41, 44 to 45 and 48 to 49) and non-commercial real estate, finance and insurance services (from NAICS 52 to 53), employment is a principal indicator for the services sector.
Services industries have become important leading indicators of economic changes. In the past decade a series of slowdowns and shocks have affected the Canadian economy, e.g., the bursting of the IT bubble, the 9/11 terrorist attack in 2001, SARS, and the 2008-09 global financial crisis. All these shocks are transmitted rapidly to the whole economy and in particular to the services sector. Policy makers recognize the importance of having a better grasp of the size of the services sector and more timely current indicators, so that they can develop and make adjustments to policy. At the sub-annual level, improved indicators of service industries' output would help to better monitor and assess current economic conditions.
What data are produced by QIRI?
The QIRI provides new sub-annual information to help monitor, on a timely basis, the performance of a subset of service industries. The focus of the program is on the tracking of the current trend of the production activity in these service industries and on the short-term evolution of operating revenue. Therefore the primary output of the program consists of quarter-to-quarter movements. Given this emphasis on movements, and specially the identification of turning points, it was decided to not publish levels (for revenue – the observed variable) at this time, but rather publish an index in current dollars of quarterly revenue movements with a base year of 2007. This simplifies the statistical process significantly and in particular removes the problem of possible confidentiality conflicts while allowing for the reconciliation to annual program levels to be completed at a later date.
How does QIRI determine which industries to survey?
Service Industries Division has the mandate to collect and disseminate estimates on a large portion of Canada's service sector. The issue of what industries to select for sub-annual output indicators involved a variety of considerations:
- System of National Accounts needs (GDP improvements, sub-annual data gaps). For example, the SNA identified sub-annual information from NAICS 8111 – Automotive repair and maintenance - as being a high priority for GST development;
- GST applicability. Research was also aimed at how the GST applied to goods and services along industrial lines. For example, with the GST, certain products are GST-exempt (exports) while others are zero-rated (basic groceries).
- Data coherence and comparability. An important factor to consider in what industries to select was internal data coherence at Statistics Canada and the ability to compare data internationally.
After appropriate consultations, the following industries in the services sector were selected to be included in the QIRI:
| NAICS | Industry |
|---|---|
| 5111 | Newspaper, periodical, book and directory publishers |
| 5312 | Offices of real estate agents and brokers |
| 5322 | Consumer goods rental |
| 5323 | General rental centres |
| 5412 | Accounting, tax preparation, bookkeeping and payroll services |
| 5413 | Architectural, engineering and related services |
| 5414 | Specialized design services |
| 5416 | Management, scientific and technical consulting services |
| 5418 | Advertising, public relations and related services |
| 5613 | Employment services |
| 5621 | Waste collection |
| 5622 | Waste treatment and disposal |
| 5629 | Remediation and other waste management services |
| 7131 | Amusement parks and arcades |
| 7139 | Other amusement and recreation |
| 7211 | Traveller accommodation |
| 8111 | Automotive repair and maintenance |
| 8112 | Electronic and Precision Equipment Repair and Maintenance |
| 8113 | Commercial and industrial machinery and equipment repair and maintenance |
| 8121 | Personal care services |
| 8122 | Funeral services |
| 8123 | Dry cleaning and laundry services |
Why are data for only some of industries available?
The purpose of QIRI was to provide timely estimates for the service sector of the Canadian economy. In conjunction with Service Industries Division's annual program, QIRI has chosen to release these industries based on several criteria, such as data quality and coherence with the annual program. QIRI is very much an ongoing initiative, and more industries will be released by this program in the future.
Will more industries be available in future?
Yes. QIRI is an ongoing initiative to provide timely estimates of industries in the services sector. As the program evolves over time, data on more industries will be made available.
What is the North American Industrial Classification System (NAICS)?
NAICS is method of classifying businesses that was developed as a partnership between various statistical agencies in Canada, the United States and Mexico. QIRI uses NAICS 2007 as its classification system. More detailed information on NAICS can be found at:
Industry classifications
What is an index?
An index shows the rate of change in a period in relation to a fixed base period. In this case, the four quarters of 2007 are equal to 400 and are used to calculate the change in operating revenue in relation to 2007. This index shows quarterly rates of change by industry.
Why are the data indexed?
QIRI data show changes in operating revenues expressed as indices. Indices are calculated from the aggregate industry revenue data, by 4-digit NAICS and by province, with the average operating revenue for the base year (2007) equal to 100. The use of an index presents a clearer view of the underlying trend and helps identify turning points in the observed industries.
Publishing only indices (as opposed to levels) simplifies the process by eliminating confidentially and dominance concerns and the requirement to benchmark to annual data. Benchmarking is the process by which a sub-annual survey series is made consistent with an annual survey series. Benchmarking is applied to reduce confusion resulting from having two different sets of numbers. It is desirable for estimates from two different surveys to be consistent and coherent, but coherence does not necessarily imply full numerical consistency. Where annual and sub-annual surveys cover the same industries, the former may target detailed measures of levels and the latter, measurement of short-term trends.
Although benchmarking would ideally be applied only to correct for sampling variability, there are conceptual, methodological and operational sources of incoherence between surveys. While conceptual and methodological differences between surveys can be removed, the need for coherence must be balanced with the need for design efficiency. Operational differences are often present for a good reason, including the need to accommodate respondents.
For these reasons, the decision was taken to not benchmark QIRI data to annual data. Instead, a reconciliation process will be undertaken at a later date between the QIRI data and the data from the annual surveys.
How are quarterly indices annualized?
To annualize quarterly indices, add the four quarters of a given year and divide by four. As 2007 is the base year, it should always equal 100, though there may be a small variance due to rounding.
How reliable are the data?
Due to the unique nature of the program, a criterion for data quality evaluation for QIRI had to be designed. Normally a survey program will evaluate the reliability of their data using the coefficients of variation of a particular data point to determine whether the data point is reliable enough to be released. As this program incorporates a census of an entire industry, either through survey or administrative data, a new method of data evaluation was required. The three criteria that determine the quality of the data are the revision rate between revised and preliminary data, the coefficient of variation due to imputation and the response rate to the census survey. A score is determined based on these criteria which are as follows:
A: Excellent
B: Very good
C: Good
D: Acceptable
E: Poor, use with caution.
These data quality indicators appear directly in the data tables available on CANSIM for each data point and can be used by the data user to determine the reliability of the data.
What are the data sources for QIRI?
The QIRI program has two major data components. For simple units (those which operate in only one province and one NAICS), only GST data are used. These tend to be smaller businesses. GST sales revenue is considered a very good proxy for operating revenue. For complex units (those which operate in more than one province and/or in more than one NAICS), a survey is used to collect information from respondents on operating revenue by province. Complex units tend to be larger businesses. Since all in-scope complex enterprises are surveyed, the survey is, in fact, a full census. These data are combined and aggregated and then converted into an index.
What is the difference between simple and complex units?
For the purposes of QIRI a simple unit is one that undertakes only one business activity at the 4-digit NAICS level and conducts this business activity in only one province. A complex unit either undertakes more than one business activity at the 4-digit NAICS and/or operates in more than one province.
Why are the data adjusted for seasonality?
Socio-economic time series can be broken down into five main components: the trend-cycle, seasonality, the trading-day effect, the Easter holiday effect and the irregular component.
The trend represents the long-term change in the series, whereas the cycle represents a smooth, quasi-periodic movement about the trend, showing a succession of growth and decline phases (e.g., the business cycle). These two components—the trend and the cycle—are estimated together, and the trend-cycle reflects the fundamental evolution of the series. The other components reflect short-term transient movements.
The seasonal component represents quarterly fluctuations that recur more or less regularly from one year to the next. Seasonal variations are caused by the direct and indirect effects of the climatic seasons, institutional factors (attributable to social conventions or administrative rules; e.g., Christmas) and technological factors.
Lastly, the irregular component includes all other more or less erratic fluctuations not taken into account in the preceding components. It is a residual that includes errors of measurement on the variable itself as well as unusual events (e.g., strikes, drought, floods, major power blackout or other unexpected events causing variations in respondents' activities).
The seasonal and irregular components conceal the fundamental trend-cycle component of the series. Seasonal adjustment (correction of seasonal variation) removes the seasonal, trading-day and Easter holiday effect components from the series, and it thus helps reveal the trend-cycle. While seasonal adjustment permits a better understanding of the underlying trend-cycle of a series, the seasonally adjusted series still contains an irregular component. Slight quarter-to-quarter variations in the seasonally adjusted series may be simple irregular movements. To get a better idea of the underlying trend, users should examine a few quarters of the seasonally adjusted series.
The QIRI series are seasonally adjusted by 4-digit NAICS and by province. National totals by 4-digit NAICS are also directly seasonally adjusted. Because of the nature of the computations involved, seasonally adjusted provincial and territorial series might not sum up to their seasonally adjusted national total. To correct these discrepancies, a reconciliation (raking) process is applied and the additivity is restored by slightly modifying the provincial/territorial series. Seasonally adjusted series are expressed as an index with the base year (2007) equal to 100. Indexing can hide the additivity of the provincial/territorial series to their national total.
Note:
Jerry Fiori, Larry McKeown & Nathalie Taktek, The growing importance of the service industries: The need for sub-annual indicators, Statistics Canada, Catalogue no. 11F0024MIE2005000, May 2005.