Accountability Employment Trade and investment Transparency

November 2013 SEPH: More on the ‘unclassified’ jobs boom

Chart 1 Unclassified business payrolls
Chart 1 - SEPH unclassified business payrolls

Source: CASNIM Tables 281-0025 and 281-0063 Employment (SEPH), seasonally adjusted, for all employees for selected industries classified using the North American Industry Classification System (NAICS), monthly (persons), Statistics Canada

According to the preliminary figures in this morning’s SEPH release, payroll jobs are down 27,600. The good news: industry-classified jobs were down only 19,100, the remaining 8,400 loss accruing to the mysterious ‘unclassified’ sector – which now accounts for ‘only’ 417,100 jobs.

As mentioned in yesterday’s Economy Lab post, the number of unclassified jobs nearly equal the job total of four broad industry sectors combined. Statscan should either stop referring to the bold-highlighted figure as the ‘sector aggregate’ or footnote that it includes hundreds of thousands of payroll jobs from so-called businesses not classified to any industry sector.

SEPH-LFS comparison

That 27,600 SEPH-reported job loss for November 2013 doesn’t compare favourably with the 21,600 LFS-reported job gain for the same month. As reported in the November 2013 LFS release, 19,100 of those 21,600 LFS jobs were of the ‘self-employed’ variety.

The occasional LFS flourish of self-employment during the Great Recession – and still-shaky recovery five years on – tends to mask the Canadian economy’s (in)ability to create jobs. The November 2013 LFS self-employed job gains were lost twice over as the following month’s LFS release reported a drop of 37,900 self-employed. This highlights self-employment during an economic downturn and shaky recovery often isn’t ‘real’ employment (data quality/reliability aside); rather, it’s households doing their best to cope with a weak labour market.


Not that this was foreseen (well, kinda) but the contrast between the November 2013 SEPH and LFS releases underscores the importance of getting the payroll survey right. Unfortunately, as Chart 1 shows, the SEPH payroll employment data’s gone wrong in recent years.

Leave a Reply

Your email address will not be published. Required fields are marked *