Skip to main content

Workers at a construction site in Montreal, Feb. 14, 2012. The last jobs report showing an 82,300 employment gain was the biggest bump since the onset of the recession and the fourth-largest increase ever. Most economists are skeptical about both the outsized gain and the job market’s weakness in previous months.

Ryan Remiorz/CP/Ryan Remiorz/CP

Bumpy patterns in Canada's labour force survey are causing headaches for economists.

The last jobs report showing a 82,300 employment gain was the biggest bump since the onset of the recession and the fourth-largest increase ever. Most economists are skeptical about both the outsized gain and the job market's weakness in previous months.

It also comes after plenty of head-scratching at the provincial level. Quebec's job market suggested the province was in recession-like conditions by the end of last year – even as other economic indicators showed modest growth. Then in March, job growth suddenly surged in the province.

Story continues below advertisement

The unevenness has sparked a flurry of questions for people like Stéfane Marion, chief economist at National Bank Financial, who uses the data to produce economic forecasts.

When he meets with foreign investors keen on Canada or provincial governments preparing budgets, one theme dominates: What is wrong with Canada's jobs market? "I've been questioned a lot," he says.

The labour force survey is the earliest major economic release of the month, and most economists view it as the most important indicator, influencing monetary policy, public policy and financial markets. Companies use employment data to budget where and when to hire, and workers use it to assess where prospects are brightest.

Last month's employment jump, after four months of little change, highlights how volatile jobs data can be, and the importance of using three or six-month trends to assess the real health of the labour market.

The labour force survey is based on answers from 56,000 Canadian households each month. Canada's sample size is almost the same as that of the United States, even though the U.S. has roughly 10 times the population.

It's a big sample, but still paints only a rough picture of employment trends. A gain of 82,000 jobs last month means there is 95-per-cent confidence that employment rose between 29,000 and 135,000 last month.

It is the most well-designed survey at Statistics Canada, painting an "extremely" accurate picture of employment at the national level, said Philip Cross, an Ottawa-based economic consultant.

Story continues below advertisement

"But it's not perfect – all data should be scrutinized from a point of view of comparing it to other data," said Mr. Cross, who is formerly the agency's chief economic analyst.

Mr. Marion offers a prescription. Bring Canada into line with the U.S., where payrolls data is published on the same day as the household survey, letting economists cross-check the data. In Canada, the survey of employment, payrolls and hours, or SEPH, is published with a two-month lag, though it is subject to far fewer revisions than that of the U.S.

"Why is it that we need to wait almost two months to get payrolls data in Canada whereas in the U.S. they produce them on the same day?" he says.

Douglas Porter, deputy chief economist at BMO Nesbitt Burns, concurs. "You would think in this day and age of technology that some of these surveys could be sped up a bit," he said.

Statscan says releasing both reports on the same day is impossible. The labour force survey takes place in the middle of the month, while SEPH data collection isn't even finished until after the end of the month. The U.S. also has a speedier system of collecting payrolls data directly from state records.

That explanation is unlikely to satisfy people like Mr. Marion, who says too much is at stake for the status quo to continue.

Story continues below advertisement

"There is a need for Statcan to prioritize its focus on the main data releases, of which employment statistics are king – even more so in a world of structural and demographic changes."

Statscan releases its February payrolls data on Thursday.

Report an error Editorial code of conduct Licensing Options
As of December 20, 2017, we have temporarily removed commenting from our articles. We hope to have this resolved by the end of January 2018. Thank you for your patience. If you are looking to give feedback on our new site, please send it along to If you want to write a letter to the editor, please forward to