Survey of Employment, Payrolls and Hours (SEPH)
Summary of changes
Activity on this program started: 1919
The Survey of Employment, Payrolls and Hours (SEPH) is now part of Statistics Canada's Integrated Business Statistics Program (IBSP), providing a more efficient model to produce economic statistics in a robust and flexible manner while reducing response burden on business respondents. Detailed information about the IBSP is available from the Behind the data module at https://www.statcan.gc.ca/eng/dai/btd/ibsp.
Following this release, the monthly portrait of job vacancy information will no longer be available through the Job Vacancy Statistics (JVS) program and will continue to be available through the Job Vacancy and Wage Survey (JVWS). Details on this change to job vacancy data can be found at http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=5202.
With this release, SEPH has started using the NAICS 2017 version 3.0 (North American Industry Classification System) instead of NAICS 2017 version 2.0.
With this release, SEPH has started using the NAICS 2017 version 2.0 (North American Industry Classification System) instead of NAICS 2012.
Statistical activity - This program tells a more complete story of current labour market events. A fifth survey, Job Vacancy and Wage Survey (JVWS), has been added. It will provide labour market demand information by occupation, offered wage and region.
Sampling - The survey is conducted on a quarterly sample of 100,000 business locations, except for the first quarter of 2015, which includes a sample of about 67,000 business locations obtained from the regular quarterly sample, or two-thirds of the regular sample. As a result, comparisons of the first quarter of 2015 data with data from subsequent quarters should be made with caution.
Sampling - In 1983, SEPH was created with a sample size of about 70,000 establishments. With the aim of reducing response burden associated with the survey, Canada Revenue Agency made an agreement with Statistics Canada to add two fields on employment and gross monthly payroll information to its payroll deductions remittance forms. These fields were first added to forms in 1993, allowing for a reduction in sample size to 29,000 units and later to 18,000 units. The sample was reduced to the current level of 15,000 establishments in 2009 during the last survey redesign.
With this release, SEPH has incorporated an additional method in assigning 2012 NAICS codes to businesses. The impact of this change is a reduction in the payroll employment level in the unclassified businesses category and slight increases in employment in most of the classified industries. Earnings data for some industries are also affected. This method has been applied back to 2008, and will be in place going forward. The unclassified category now accounts for 1% to 2% of payroll employment compared with 2% to 3% before the change.
At the same time, seasonally adjusted data have been revised based on the latest seasonal factors. Historical revisions have also been made to a small number of industries by province or territory. These data were revised back to 2001.
In addition, quality indicators have been added to CANSIM table 281-0039.
With this release, SEPH has started using the 2012 North American Industry Classification System (NAICS) instead of NAICS 2007. Data were revised historically back to 1991. As well, seasonally adjusted data were revised based on the latest seasonal factors. Historical revisions were also made to a small number of industries by province or territory. These changes have little impact on SEPH estimates.
Statistical activity - This program tells a more complete story of current labour market events. A fourth survey, Job Vacancy Statistics (JVS), has been added. JVS offers information on labour demand by reporting on the number of job vacancies by industry.
As of this reference period, a new seasonal adjustment model (X-12-ARIMA) has been introduced.
Collection instrument - Starting with the January 2011 reference month, two questions were added to the Business Payroll Survey (BPS) to collect data on job vacancies (see record number 5202). These questions were: Did you have any vacant positions on the last business day of the month, and how many?
Data on job vacancies are based on three-month moving averages. For example, data for September are based on an average of the data from July, August and September. Data will start being released for September 2011.
Estimation - With the release of January 2009 SEPH data, an improved estimation method for earnings and hours data was introduced and estimates back to 2001 were revised to ensure continuity in data series.
Starting with the release of data for the January 2008 reference month, current and historical estimates from the Survey of Employee Payroll and Hours (SEPH) are using the 2007 North American Industry Classification System (NAICS) instead of NAICS 2002. While the differences between the classification systems are minor, the greatest changes are within the sectors of; information and cultural industries (NAICS 51) and in finance and insurance (NAICS 52). Estimates have been revised back to 1991 to account for the change to NAICS 2007. As per the normal SEPH annual revision cycle, historical corrections to other series are also available.
Starting with the release of data for the January 2004 reference month, estimates are published based on the 2002 version of North American Industry Classification System (NAICS). This change only affects estimates for industry groupings within the Construction (NAICS 23) and Information and Culture (NAICS 51) industries. In addition to the switch to NAICS 2002, there is a slight change to the definitions for groupings "durable and non-durable goods" of manufacturing.
Starting with the release of data for the January 2001 reference month, estimates were published based on the 1997 North American Industry Classification System (NAICS). Previously, estimates were based on the Standard Industrial Classification (SIC) of 1980.
Statistical activity - Together, three monthly surveys tell a more complete story of current labour market events. These surveys are: the Labour Force Survey (LFS), the Survey of Employment, Payrolls and Hours (SEPH) and the Employment Insurance Statistics (EIS).