Employment Minister Jason Kenney continued to tout online job advertisements as a great way to measure Canada's job vacancy rate even though he had been warned of their unreliability, an internal memo shows.
The briefing note, obtained by PressProgress under access to information law, was drafted by departmental bureaucrats following the release of the federal budget. Kenney had asked for the information after the "media mentioned the existence of two different measures of the job vacancy rate that present conflicting messages about the current state of labour market conditions in Canada," states the briefing note, delivered to his office on February 27, 2014.
The memo explained, in part, that "the upward trend in the number of online job postings over time reflects the growing adoption of the Internet by employers as the preferred tool to advertise job openings, and the improvements made by Wanted Technologies to increase coverage of online posting boards."
A few weeks earlier, the Conservatives published a jobs report alongside the federal budget saying the job vacancy rate had been "increasing steadily" since 2009 and that they expected it to continue to rise. The analysis, conducted by Wanted Technologies, was based on online job postings (and conveniently supported the Harper government's explanation to justify the growth of the Temporary Foreign Worker Program under its watch).
Following the release of the jobs report, Statistics Canada released its own report contradicting it. The agency reported the number of job vacancies was the lowest recorded (under 200,000) since Statistics Canada began tracking the figure in March 2011. Statistics Canada based its analysis on its Business Payroll Survey of 15,000 establishments.
Kenney's public response to the contradictory findings?
Throughout the spring, he defended the jobs report and the use of online job postings to measure job vacancy rates. He persisted even after economists explained that the rising rate touted by the government (and based on online job postings) evaporated if you excluded the online marketplace Kijiji from the mix of online sources.
Privately, bureaucrats had explained to Kenney that, overall, using online job advertisements to measure job vacancy rates had serious limitations. In other words, it wasn't just a Kijiji problem.
Here are some highlights from the briefing note, which explains why it's a bad idea to rely on online job ads to measure job vacancy rates. Although it's tempered by bureaucrat-speak, the note could be called Why Statistics Canada gets it right.
On Statistics Canada's approach:
- "To be considered as a vacant position in the Statistics Canada survey, the position must exist and be unfilled on the last business day of the month, the employer must actively be seeking workers from outside the organization and the work may commence within 30 days. This definition of a vacant job is similar to the one used by national statistical agencies in other countries."
- "Of note, although the US economy is much larger than the Canadian economy, the size of the business sample used by Statistics Canada (15,000 establishments) is similar to the one used in the US survey (16,000 establishments)."
- "Surveyed employers receive the same questionnaire from Statistics Canada, with the same information about what vacancies to report. This is to ensure consistency and comparability of information across employers, sectors and regions."
On using online job postings to determine a country's job vacancy rate:
- "Practices regarding the use of online job postings can vary markedly among employers, sectors and regions."
- "Unlike the job vacancy data collected by Statistics Canada, job advertisements cover more than hiring needs related to existing vacant positions. Online job postings may reflect anticipated future hiring needs, with current advertisements serving to build lists of potential candidates for positions that will become vacant or be created much later."
- "Also, online job postings can capture existing unfilled positions in which work may not start for several months (indeed many positions do not specify a starting date for the position). For such reasons, online job postings constitute a proxy rather than a direct measure of current hiring needs."
- "Also, the upward trend in the number of online job postings over time reflects the growing adoption of the Internet by employers as the preferred tool to advertise job openings, and the improvements made by Wanted Technologies to increase coverage of online posting boards."
- "It should be noted that different monthly estimates of the number of online job postings can be obtained from the daily data released by Wanted."
On what that looks like:
Here's the full briefing note
Photo: Mostly Conservative. Used under a Creative Commons BY-NC-ND 2.0 licence.