The pandemic has thrown a wrench into the statistical machinery that tracks job numbers
Interpreting and comparing employment stats demands a lot of care right now, writes Miles Corak
This column is an opinion by Miles Corak, a professor of economics at The Graduate Center, City University of New York. He has worked for Statistics Canada, Employment and Social Development Canada, and served in 2014-15 as a member of the Economic Council of Advisors to the Liberal Party of Canada. For more information about CBC's Opinion section, please see the FAQ.
Erin O'Toole, with a sense of indignation and urgency, has boldly proclaimed that "We lead the G7 in unemployment." Statistics in the service of partisan politics are often, to put it gently, rather elastic in their meaning, so it is natural to wonder: do we really lead the pack in the dubious distinction of having the highest rate of unemployment?
Statistics Canada reported Friday that the unemployment rate stands at 8.2 per cent, a full two percentage points above that of the United States. As opposition leader O'Toole points out by citing the Organization for Economic Cooperation and Development (OECD), that's higher than in many other rich countries.
But more care is needed to uncover the true meaning of these numbers, because the pandemic has twisted the workings of the statistical machinery that in normal times serves us well.
In fact, the OECD suggests that many comparisons be taken off the table. In a note on Employment and unemployment statistics during the COVID-19 crisis, it cautioned that, "the unprecedented impact of COVID-19 … affects the cross-country comparability of unemployment statistics."
In many European countries, people who have been temporarily laid off — those who have lost their jobs but have an expectation of being recalled — are counted as employed.
The Canadian and American statistical agencies do just the opposite, counting them as unemployed. This difference in methods usually doesn't amount to much, except of course during the pandemic, when the number of temporarily laid-off shot to the stratosphere.
In other words, the OECD is saying that its off-the-shelf statistics currently come with a warning label: use with caution, particularly across the Atlantic divide. So, claiming that Canada stands worst in the G7 is likely an apples-to-oranges comparison.
The other comparison that beckons is to the United States. The suggestion has been made that Canada scores poorly when compared to the unemployment rate south of the border.
The statistical agencies in the two countries use a number of rules of thumb to classify someone as "unemployed." It is not just about being without a job. The unemployed are those without paid work, and who at some point in the past month have also looked for work.
But there is a consequential difference in how this official "unemployed" classification is determined.
The U.S. Bureau of Labor Statistics requires an American without paid work to do something to look for work, but it must be something that in principle would lead to a job offer. Directly contacting an employer or attending a job interview are among the activities that will do the trick.
Statistics Canada casts the net more widely to also capture those who just look at want-ads, check with a friend or relative, or engage in other activities that on their own wouldn't lead to a concrete offer.
If Statistics Canada used the same measuring rod as its American counterpart, the Canadian unemployment rate would be almost one and half percentage points lower, cutting the two-percentage point gap between the two countries by more than half.
It is hard to see urgency in these numbers, as this gap, however measured, is not much different now than during 2015, or for that matter 2005.
Besides, Americans are actually questioning their official unemployment rate. Jerome Powell, the head of the Central Bank and arguably one of the world's most powerful economic policy makers, has recently stressed that, "published unemployment rates during COVID have dramatically understated the deterioration in the labour market."
In part this is because the U.S. Bureau of Labor Statistics' methods are not well suited to pandemic times, leading many jobless people to be misclassified as employed.
However, he also notes that millions of Americans say the pandemic has prevented them from looking for work. So even if they very much want to work, they're not officially unemployed under the U.S. rules because they're not actively looking — some because no jobs are available, others because of worries about the virus, and others for child-care reasons.
Some economists have made calculations that suggest the "realistic unemployment rate" in the U.S. is 8.2 per cent. In Powell's view, the official unemployment rate, now standing at 6.2 per cent, should be upped to almost 10 per cent.
These adjustments put the American number well within the range of the Canadian, implying that there is not much difference between the rate in the two countries.
All this said, if the concern is about jobs, then the Canada-U.S. scorecard should be based on employment rates. This might be an easier apples-to-apples comparison.
In this case, the cross-country patterns are not much different, and indeed there's been more improvement in Canada. Employment has certainly tanked, falling significantly and to the same degree in both countries, but the Canadian bounce-back from the lows of last April has been quicker.
The bottom line?
Canadians on average have been more hopeful that a job search will pay off, more likely and able to be looking for work, and hence more likely to be classified as unemployed than Americans. This, along with an understatement of the American unemployment rate, creates the illusion that Canada is doing worse.
Not at all the message the opposition leader has been sending, but somewhat closer to the true meaning of the statistics, elastic as that may be. The pandemic has muddied the statistical waters and demands more care in interpreting numbers that are usually easy to understand.