Canada’s economy is booming… or not. Depends on whether you’re in economics or finance—or in academia or looking for a job. Statistics Canada (Stat Can) seasonally adjusted data shows the country added tens of thousands of jobs in January, pushing the unemployment rate lower. Unadjusted data tells a very different story, with the country adding nearly 200k jobseekers, pushing the unemployment rate to the second highest monthly print since the 2021 recession. We promise, you’ll understand jobs data and how to read it better than your local politician in just a few minutes.
Canada Adds 76k Jobs… Sort of. Depends How It’s Counted
Canadian headline employment data is showing strong improvements. The country added 76k seasonally adjusted jobs in January, helping to push the unemployment rate down to 6.6% (-0.1 points). The rate has shown a 0.3 point improvement since peaking in November 2024, but remains 1.6 points higher than March 2023. It’s been a remarkable turnaround and policymakers seem to be eager to test this data at the polls.
Find the numbers hard to believe? That’s become a fairly common issue, which policymakers refer to as a “vibecession.” The economy isn’t bad, but people feel like the data is wrong because they’re just in a negative mindset. Unfortunately, it’s a little more complicated than that. Let’s talk about seasonal adjustments.
Seasonally Adjusted Affective Disorder (SAAD)
Seasonal adjustments are a statistical method to remove predictable patterns from time series data. Spending is very seasonal, especially when it comes to large, expensive items. People buy homes and cars in the Spring, don’t buy winter coats in July, or air conditioners in January. Adjusting for these predictable patterns helps to remove noise and make it easier to spot trends over the long term.
Predictability is the key word there, especially as the world becomes more or less predictable in different areas. The economy isn’t as dependent on agriculture as it was 100 years ago when the idea was first introduced, and modern advancements in tech mean the road is blown so you can get to the factory. The future is now.
At the same time, the volatility of trade is less predictable than it once was. Some rowdy guys in wigs would throw the tea in the harbour once every 300 years back then. Now an election in a single country can suddenly re-write how trade is conducted. Economists in government and academia still consider seasonal adjustments as the de-facto standard, but it’s less relevant in finance.
“You can’t seasonally adjust a terrorist attack,” said JP Morgan economist Marc M Goloven back in 2002.
Goloven was dismissing the “false positives” seen on Wall St a year after the 9/11 terror attacks. Global consumption did not have a predictable pattern, and consequently the adjustments made data look much better than they were in reality. There goes predictability.
“Seasonally adjust me, I’m George Clooney,” explained Wells Fargo economist Mark Vitner, who could only barely pass as the actor’s twin. He was dismissing eroding job data during the 2019 economic slowdown that became overshadowed by 2020.
The skepticism isn’t just a one-off issue either. The EU warned internally how little research on seasonal adjustments exists, despite the dogma surrounding it. The economy there is probably fine after dismissing the issue… does the EU still exist?
We know. What if our data doesn’t like techno and soccer? Well, the US Federal Reserve recently warned the events of 2020 would distort data in the coming years. An issue the central bank is familiar with, after realizing post-Great Recession data indicated a recovery occurred much faster than believed, resulting in delivering the wrong policy.
None of this is to imply that seasonal adjustments are an intentional distortion or conspiracy. It’s just an issue of understanding it may not provide the data a person is looking for, and it’s not just because they’re vibin’ wrong. However, unadjusted data is messy looking and seasonally adjusted data makes it easier to spot a trend—even if it’s the wrong trend. So there’s that.
On that note, let’s look at the unadjusted employment model. The unadjusted data makes more sense to the average person who needs a job to pay their bills. It probably doesn’t go over well when a person gets fired and applies for a mortgage with their income stated as seasonally adjusted. Maybe at some Canadian banks…
Canada Actually Lost 174k Jobs & Gained 198k Unemployed People
The unadjusted employment data isn’t the worst Canada has seen, but it’s certainly less than flattering. Seasonally adjusted the country added jobs, but unadjusted data shows 173.5k fewer jobs in January. Last month is usually a big month for joblosses, but this was bigger than usual.
More important is the unemployment rate, which only includes people ready, willing, and able to work. That means no full-time students, or people who plan to look for a job next Tuesday but never get around to it. Canada’s unadjusted unemployment rate climbed 0.9 points to 7.1% in January. With the exception of August 2024, this would be the highest unadjusted unemployment rate since 2021.
Canadian Unadjusted Unemployment Rate Hits 2nd Highest Month Since 2021 Recession
The seasonally adjusted and unadjusted unemployment rate across Canada.
Source: Stat Can; Better Dwelling.
Not exactly something to dismiss simply because it’s smoothed out in adjustments. There’s also the issue that the adjusted rate is 7% lower than unadjusted in January 2025, but it was only 6.6% in 2024. In 2017, the rate was only 5.5% lower for the month. The seasons are creating much more noise these days compared to 8 years ago. Maybe it’s just climate change.
Ultimately the two data points are important to different people for different reasons. A bureaucrat or journalist looking for some good news, may prefer the seasonally adjusted unemployed population fell 14.9k to 1.49 million people in January. Those looking for a job are likely more concerned that the unadjusted data shows the unemployed population rose 198.1k to 1.57 million unemployed people searching for work. In academia, the people removed by seasonal adjustments don’t exist—but in reality, more people are competing for roles.
There’s also the issue of those not looking for a job. As discussed last week, Canada massively overinflated the share of immigrants it was able to retain. Not just over the past few years, but this was a multi-decade issue. That portion of the population is modeled into the non-participation rate, downplaying the number of people able to exist without employment. On the upside, the millions of people that only exist in estimates means the housing shortage is really bad in cities, but not as bad as presented. In fact, the people fleeing cities like Toronto likely caused a bigger shortage than they were told Toronto had. Unfortunately prices are based on expectations, since reality can just be adjusted in an academic model.