To keep credibility, Statistics Canada should explain Quebec’s bogus job slump
Remember the huge recession in Quebec late last year, the one in which the province suffered its biggest three-month loss of employment ever?
No? Neither do most of us, even though Statistics Canada’s official employment reports at the time said Quebec lost a stunning 69,500 jobs from October through December of last year. This, understandably, caused deep consternation among Quebec’s political leaders.
Now it’s increasingly clear that the slump never happened. This week a less timely Statscan report, which counts up those on Canadian payrolls, finally showed that, far from plunging, Quebec payrolls actually grew right through the end of the year.
While most economists still respect Statistics Canada’s professionalism, there’s increasing concern that without some explanation what went wrong, the credibility of its official household survey of employment, a key economic barometer, will be damaged.
“I rely a lot on the employment numbers as a starting point for my forecast and my view of the economy. If it’s not reliable, then that’s not good,” says Carlos Leitao, chief economist at Laurentian Bank Securities. And in his mind, the new payroll numbers confirm what he already strongly suspected: that the official job survey “is a little wacky.”
This issue has been building for months. Statistics Canada’s household survey showed the first hint of a slump in Quebec’s job market in October 2011, reporting the loss of 13,300 jobs. This got much worse in the following months. This phenom- enon wasn’t a mere blip, notes Douglas Porter, deputy chief economist at BMO Capital Markets. It lasted for three months “and was a frightening decline, the biggest one ever seen” in Quebec. But with the new payroll numbers once again contradicting the trend that was reported in the official report, economists are penning devastating comments on the inaccuracy of the official figures.
“What is really going on in employment?” asked a note to clients yesterday from Porter.
His answer: the official numbers, taken from a sampling of Canadian households and then extrapolated to the whole population, don’t make sense, since no other evidence agrees with them.
Porter finds that numbers from the payroll survey, “seem much more realistic.” It shows absolutely no evidence of a Quebec slump. Indeed, it shows Quebec employment growing right through the end of last year.
Laval University economist Stephen Gordon is even more blunt. In a blog posting entitled “An update on the apparently nonexistent Quebec recession,” he shows in great detail how every economic indicator other than the official job survey shows Quebec’s economy growing normally.
Some telling examples: Quebec’s employment insurance claims didn’t go up in this period compared with 2010. And retail sales, including those of autos, actually grew faster than in the rest of Canada.
It’s not just Quebec whose numbers look questionable when you compare them to Statscan’s count of payrolls, Porter notes. The Maritime provinces suffered a mirror image of Quebec’s apparent distortion: while the official numbers showed employment stable or up, the payroll survey showed losses in three of the four provinces.
Indeed, while the official figures showed Quebec employment shrinking by 1.3 per cent during 2011, while the payroll survey showed it growing by 1.1 per cent, it was Nova Scotia that had the most striking gap. There, the official figures showed a gain of 2.3 per cent, while the payroll data showed a loss of 2.1 per cent: a gaping difference of 4.4 percentage points.
It’s possible that self-employment, a category left out of the payroll survey, accounts for some of the disagreement, but probably not much.
For example, even when you remove self-employment from Quebec’s official fourthquarter figures, they still show a huge loss of 54,000 jobs. The payroll survey shows a gain of 5,900.
Maybe it’s time for an explanation.