Construction workers toil inside the Barron Building as crews convert it from office space to residential apartments in Calgary on Dec. 12, 2023.Jeff McIntosh/The Canadian Press
Friday’s Canadian employment report – the last for 2023 – wasn’t great. It also wasn’t awful. Mostly, it wasn’t particularly clear.
The report showed no job growth for December. Yet it also showed no increase in the unemployment rate.
Full-time work declined by 24,000 positions, which was made up for by a near-identical increase in part-time work. Yet the total number of hours worked rose by a healthy 0.4 per cent from November, despite a strike of Quebec public-service employees in December.
Critically, average hourly wage growth surged to 5.4 per cent year over year, from 4.8 per cent in November. That’s a big step in the wrong direction for anyone connecting the dots between the labour market and inflation (I’m looking at you, Bank of Canada).
Still, make no mistake, this most certainly is a softening job market. Don’t let the contradictions in the December data throw you off course. It’s the broader trend that matters – and that trend is toward a loosening of labour conditions.
With the book now closed for 2023, the totals would, at a glance, suggest that the labour market had a pretty good year. Total employment rose by 430,000; that’s a total that, prior to the COVID-19 pandemic, had been topped only twice in two decades. Wage growth was the strongest since 2006.
But a closer look at how things have evolved over the past 12 months tells a very different story.
The employment rate (the percentage of Canadians aged 15 and older who have a job) has been in decline since last spring, and it continued that decline in December; it’s now the lowest it has been in nearly two years. In the past six months, the working-age population has increased by a half-million, yet the national job count has risen by just 140,000.
For many economists, the employment rate is a much more valuable indicator of labour market strength than the more familiar unemployment rate; it points to the share of the entire pool of labour that is being used. A shrinking employment rate is a vital indicator both of growing slack in the labour market, and a sluggish economy more generally.
Meanwhile, the rise in the unemployment rate since spring has been widely noted. Unemployment was unchanged in December, at 5.8 per cent; but it has been rising steadily since last spring, when the rate was just 5 per cent.
The December lull looks likely to be a brief statistical pause in a continuation of that rising unemployment trend in 2024. Economists at the country’s six big banks forecast that the unemployment rate will rise, on average, to 6.6 per cent by the end of this year.
So, what should we make of the inconsistencies within Statistics Canada’s December report?
Every once in a while, we get a head-scratching monthly release that reminds us that these data are derived from a survey of Canadian households. Like any survey, there are meaningful margins of error that can, sometimes, spit out inaccurate and misleading results.
As Statscan explains, its labour force survey covers a single week within each month, in this case Dec. 3-9. About 48,000 households – some 96,000 people 15 or older – were asked about their employment status. That’s 0.3 per cent of the working-age population, whose answers are relied on to represent the whole.
Statscan states that, statistically speaking, December’s estimated monthly employment total is considered accurate to within plus or minus 30,700 jobs, 68 per cent of the time. If you want to increase that to a 95-per-cent confidence level, then the range is plus or minus 61,400 jobs.
The unemployment rate itself has an accuracy of plus or minus 0.1 percentage point at the 68-per-cent confidence level, and 0.2 percentage point with 95-per-cent confidence.
Easy to see, then, how a single survey could produce a substantial statistical head fake. As Statscan cautions in the report itself, “monthly estimates will show more variability than trends observed over longer time periods.”
Still, the increase in the annual wage-growth rate was large enough to raise eyebrows. As the Bank of Canada considers interest-rate cuts in 2024, it has been adamant that it needs to see downward momentum in wage growth (among other things) to be convinced that inflation is headed sustainably toward the bank’s 2-per-cent target. The December jump may be a one-off, but the bank won’t breathe easier until it sees that upward blip reversed.
Yet it’s notable that average wages actually declined in December, 2022, the month against which the December wages were compared to calculate the rate of increase. That certainly exaggerated the size of the increase.
Within the December data, there may be some signs of easing wage pressures to come. For example, the goods-producing industries – which led the way in wage gains in 2023, with an average of 6.2 per cent – are also the sectors showing the sharpest slowdown in employment. Goods employment actually fell by more than 200,000 jobs over the second half of 2023. The implication is that, for these workers, the labour supply-and-demand dynamics, and hence bargaining power, have deteriorated rapidly.
It’s another reason to believe that the path still leads to lower interest rates. Don’t let one hazy employment report distract you from the trend.
Editor’s note: A previous version of this story said Statscan’s labour force survey interviewed 48,000 individuals; this version has been corrected.