Employment Social security Youth

December 2013 LFS: An alternative view. Just because it’s the first release of the new year

Chart 1 – Real vs nominal employment
Chart 1 Real vs nominal employment December 2013

Source: CANSIM Table 282-0087 Labour force survey estimates (LFS), by sex and age group, seasonally adjusted and unadjusted monthly (persons unless otherwise noted), Statistics Canada

Told ya so (see last month’s comment re self-employment boom/busts).

Seeing as it’s been some time since this ‘alternative’ Labour Force Survey (LFS) report was started, it’s probably worth going back to update a few of the charts and tables previously presented. Unlike StatsCan, we’ve tried not to bore readers with the same two charts each month.

No place better to start than the beginning, the ‘alternative’ LFS report for July 2013. Apparently that analysis was well-received – see recent similar charts by Stanford, Corak, as well as this CCPA post).

As noted in that first post, readers relate better to job numbers than participation rates, so the loss in employment since The Great Recession was simply converted to a job figure. All the LFS employment figures are estimates adjusted every which way, so estimating a job figure that would get us back to where we started (Q3 2008) after accounting for population growth made as much sense as anything else.

Many economists are reluctant to take this approach, however, for one reason: adjusting for total population (age 15+) growth would include an increasing share of seniors over 65, the age typically associated with retirement. We didn’t think that was a major problem then, nor do we now, given the alternatives.

The LFS estimates seniors age 65+ participation rate rose from 10.1% to 13.0% from 2008 to 2013; their employment rate also rose over the same period, from 9.8% to 12.5%. It seems those CPP/OAS/GIS clawbacks to early retirement, inducements to retirement postponement (past age 70) and repeals of mandatory retirement laws have had the desired, if not woefully short-sighted, effect. That said, the LFS estimates population of seniors age 65+ increased 19.3% since 2008.

The better alternative to adjusting for population growth would be to adjust for the growth in the number of Canadians who were either looking for work (the unemployed and involuntary part-timers) or wanted work (the marginally attached). Unfortunately, StatsCan makes that impossible to do with any precision, as the LFS statistically retires seniors who aren’t in the labour force. Apparently, despite their rapidly rising labour market participation, there are no discouraged job-seekers age 65+.

While it’s not clear from the LFS questionnaire, StatsCan also excludes most full-time students. They’re neither counted as unemployed if they’re actively looking for work, nor discouraged if they wanted a job but gave up looking. However, they are counted as employed if they’re also working while pursuing their full-time studies. That’s a significant exception, considering full-time post-secondary enrolment for youth age 15-24 between 2007/8 and 2011/12 (latest year available) rose 130 thousand, up 13.3%. For context, their LFS-estimated total population rose just 60 thousand, up 1.4% between 2007 and 2011. Youth employment declined 145 thousand, down 5.5%, over the same period. While correlation isn’t causation, the simultaneous loss in youth employment and increase in full-time enrolment, both disproportionately greater than their relative increase in population, suggests many young people remained in / returned to school as they became disemployed – a form of labour market attachment.

The second best alternative to adjusting for the total number of Canadians looking for work would be to adjust for population growth less the share that selectively opted out of the labour market. Here again, the LFS data – makes it impossible to do with any precision. While the LFS questionnaire asks the questions (LFI_Q105, LFI_Q131) that would allow StatsCan to figure out if respondents retired since the previous month’s survey, it only publishes the data aggregated for all those who reported retiring over the past year. This may be a data quality issue (sample capturing too few respondents who just happen to retire during their six month rotation in the LFS). Also, this wouldn’t account for seniors re-entering the labour force.

(Likewise, full-time students aren’t asked if they selectively opted out of the labour force to return to their studies as a result of disemployment and/or discouragement.)

All things considered, until there’s a better way of doing it, simply adjusting for population growth makes as much sense as anything else.



Leave a Reply

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