Canada’s labour recovery from the COVID-19 pandemic has a steep path to climb, if little publicized – but widely trusted – data is any indication.
Statistics Canada published its Survey of Employment, Payroll and Hours (SEPH) Thursday, showing that employment rose by 44,200 in December. That report focuses on the number of employees receiving pay or benefits, and notably excludes the self-employed. Despite December’s gain, total employment was down 6.2 per cent from February, 2020.
That’s a much weaker recovery than Canada’s other key jobs report is suggesting. As of December, Statscan’s Labour Force Survey (LFS) showed that total employment was down 3.4 per cent. When the self-employed were removed, there was a drop of 2.7 per cent.
The discrepancy raises a host of questions, but few easy answers. Still, the SEPH report presents a less optimistic view of the labour recovery – and a much bigger jobs gap to fill.
“It’s another data point that we should consider,” said Brendon Bernard, economist at job-listing site Indeed Canada. “The story [with SEPH] is that there’s a bit further to go to recovery.”
The LFS tends to get the lion’s share of attention from economists and investors. A survey of households, it’s the first major release on Canada’s monthly economic calendar, and produces such widely cited figures as the unemployment rate. The results hold enough power to move the Canadian dollar and bond yields.
By contrast, the SEPH is released about seven weeks after the household survey, and thus tends to get overlooked. However, many bank economists view it as a more credible gauge of employment. In large part, that’s because of the data being used. Rather than relying on household responses, SEPH estimates are derived from multiple sources, including payroll deduction records provided by the Canada Revenue Agency.
During the first two months of the pandemic, the labour reports showed roughly similar declines in employment. Then, in May, things changed. The household survey showed employment was starting to rise again, which caught experts by surprise. Weeks later, the payroll survey pointed to another sizable round of layoffs in May.
It’s not unusual for the surveys to move in opposite directions in a given month. And over the summer and early fall, both reports showed rising employment. But since May, SEPH’s version of the labour recovery has consistently lagged behind.
One potential explanation is that the SEPH measures employment by the number of paycheques, whereas the LFS measures the number of people employed. Thus, if someone with multiple jobs loses one of their positions, SEPH registers a decline in employment, while the LFS doesn’t.
“That could be contributing to the gap,” Mr. Bernard said, although he’s unsure if that fully explains it.
A more troubling – though uncertain – explanation is the nature of surveying itself.
Statscan produces both labour reports, and it has struggled during the pandemic to survey households. The LFS response rate has hovered around 70 per cent for months, well down from an average of 87 per cent in 2019. For health reasons, the statistical agency has stopped doing in-person interviews, relying instead on phone and online collection.
While Statscan has stood by the quality of its LFS data, the concern among some economists is that people who aren’t responding have been affected differently by the pandemic. Put another way, with the LFS showing a stronger recovery, the worry is that people who were less likely to lose their jobs were more likely to respond.
“We have no idea how big of an issue it is,” Mr. Bernard said. “It’s an open question whether [low LFS response rates] help explain the gap.”
Mr. Bernard noted that other labour metrics – such as the employment rate and the number of hours worked – will be critical in assessing the country’s success in getting people back to work in the months ahead.
Your time is valuable. Have the Top Business Headlines newsletter conveniently delivered to your inbox in the morning or evening. Sign up today.