3 Canadian Labour Market The Roots of Budding Change Benjamin Tal and Nick Exarhos Chart 1 It Doesn't Look Good The highly volatile monthly job creation figures and an unemployment rate that sometimes masks more than it reveals get all the attention. But the real tale of the Canadian labour market is written far away from the spotlights, closer to where the details reside. And there, the emerging picture is of a job market that is fundamentally changing. Canadian employment dances increasingly to the tune of structural forces and less to reversible cyclical dynamics. And its not only about demographics. Job market mismatches, sticky long-term unemployment, diverging bargaining power, rising entry barriers and increased job tenure and job stability for those who clear the bar, all suggest that monetary policy aimed at the cyclical component of employment slack is aiming at a shrinking target. Productivity Catching Up Its not news that the Canadian labour market hasnt lived up to recent expectations. Job gains have trailed the pace that economic growth would have suggested. Indeed, since the beginning of 2013 the pace of monthly employment growth has been on a declining trend, with sizable drops within the past few months depressing the six-month average signicantly (Chart 1). To be sure, there are some factors that would explain why hiring would lag behind what output would imply. Real output per worker is still below its long-run trend (Chart 2, left), and a convergence there would imply a period where the gains in the economy outpace commensurate gains in employment. The acceleration in Canadian productivity (Chart 2, right) is thus a by-product of that catch-up. Falling Participation RateIts All Demographics Some believe that demographics can explain everything. In the case of the declining labour market participation rate in Canada, they are right. When referring to the impact of the aging population on labour market activity we should start using a present tenseit is already happening. Though participation rates in both Canada and the US have fallen since the onset of the recession, the rate of decline north of the border hasnt been as dramatic (Chart 3, left). And the aforementioned demographic effects have been the primary driver of that deterioration in Canada, as opposed to the prevalence of cyclical factors at work stateside. Looking at the difference between the Source: Statistics Canada, CIBC Chart 2 Below-Trend Real Output Per Worker (L) Explains Productivity Pick-up (R) Source: Statistics Canada, CIBC Canadian Employment Growth (m/m change) 0 5 10 15 20 25 30 35 12 13 14 000s, 6-mos moving avg 80 82 84 86 88 90 92 94 96 98 100 94 Q1 99 Q3 05 Q1 10 Q3 Real Output per Worker (000s) Long-Run Trend 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2006-09 2010-12 2013- 14Q1 Average Quarterly Change in Productivity Index (SAAR) % CIBC WORLD MARKETS INC. Economic Insights - June 18, 2014 4 alternative measures of unemployment in the US and Canada 1 suggests that Canada has not built up a notable overhang of discouraged and under-employed individuals (Chart 3, right). Indeed, if we were to take the demographics prevalent at the beginning of 2008 and hold them steady as age- specic participation rates varied over time in Canada, we would have seen the broad participation rate tick higher (Chart 4, left). But because the employment share gains of those aged over 55whose participation rate is much lower than younger groupshave been at the expense of prime-aged workers and youths (Chart 4, right), the participation rate has steadily crept lower. In fact, that drop in participation understates some deeper deterioration, as a growing number of Canadians aged 55 and over still in the labour market are rapidly reducing their level of job-market engagement. Since 2007 the number of older self-employed individuals has risen much faster than in any other age group, seeing their share in total self-employment rise strongly (Chart 5). Vacancies and Unemployment: Not Dancing to the Same Tune Wi th demographi cs sl owl y cl awi ng away at the employable base of Canadians, another potential structural factor may be chipping away at it further. The Chart 4 An Aging Workforce (L) Primary Driver of Participation Rate Slide (R) Source: Statistics Canada, CIBC -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 15-24 25-54 55+ Change in Share of Working Age Population Since 2008 (%) 66.0 66.5 67.0 67.5 68.0 68.5 08Q1 10Q1 12Q1 14Q1 Participation Rate Actual With '08 Demographics Chart 6 Positively Correlated Vacancy and Unemployment Rates Implies Some Labour Market Mismatch Source: CIBC calculations based on Statistics Canada's tabulations Chart 3 A Smaller Drop in Participation Rate (L) as Cyclical Forces Less Important in Cdas Participation Slack (R) Source: Statistics Canada, CIBC 60 61 62 63 64 65 66 67 68 Canada USA Pre-2008 Avg Latest Participation Rate (%) 2.5 3.5 4.5 5.5 6.5 7.5 00 02 04 06 08 10 12 14 US Canada Broader Minus Narrower Unemployment Definition Residual (%) Source: Statistics Canada, CIBC Chart 5 Self-employment Rising Fast Among Workers Aged 55+ Share of age 55+ in Self-Employment 0.10 0.12 0.14 0.16 0.18 0.20 0.22 0.24 0.26 07 08 09 10 11 12 13 14 Growth in Self-Employment (2007-2014) 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 15-24 25-44 45-54 55+ Age avg Jan-May 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 6.8 7.0 7.2 7.4 7.6 7.8 Unemployment rate (%) V a c a n c y
r a t e
( % ) Beveridge curve March 2011-March 2014 CIBC WORLD MARKETS INC. Economic Insights - June 18, 2014 5 Chart 8 Wage Gains Highest Among High-paying Industries Source: Statistics Canada, CIBC Chart 7 Inow into Unemployment Slowing (L), Long-term Unemployment Remains Sticky (R) 2 3 4 5 6 7 00 02 04 06 08 10 12 14 %, 6-mo moving avg Duration of <13 weeks as a share of Total Labour Force 0 5 10 15 20 25 30 00 02 04 06 08 10 12 14 %, 6-mo moving avg Duration of 27 weeks+ as a share of Total Unemployment Source: CIBC calculations based on Statistics Canada's tabulations 90 95 100 105 110 115 120 125 130 135 05 06 07 08 09 10 11 12 13 14 Low-paying industries Mid-paying industries High-paying industries Wage index 2005=100 3-mos moving avg High-paying Industries Led the Way in Recent Years abnormal relationship between recent vacancy rates and unemployment suggests that large swaths of those unemployed are not what employers are seeking. The Beveridge Curve should be downward sloping, i.e. lower levels of unemployment generally occur when there are higher levels of job vacancies. That intuitive relationship has failed to hold true in Canada since 2011, with higher vacancies and higher unemployment positively correlated (Chart 6) 2 . A disconnect between the types of workers desired and those that are available in the ranks of the unemployed would explain how a growing number of unlled vacancies could co-exist with a higher level of unemployedand potentially unemployableindividuals. In that context, the actual slack in the labour market would in fact be lower than what is represented in the headline unemployment rate . Diverging Bargaining Power Traces of potentially unemployable workers can also be seen when comparing inows into unemployment (becoming unemployed) and the levels of the long-term unemployed (staying unemployed). The share of those unemployed for roughly three months or less has been plumbing cycle lows recently, and is even close to all-time troughs (Chart 7, left). But the number of those who have been unemployed for 27 weeks and longer remained at elevated levels (Chart 7, right). Therefore, the sticky unemployment rate of the past two years or so is largely due to stagnation in long- term unemployment as opposed to an increase in the number of newly unemployed. And the wage mechanism reects this reality clearly. Unlike past recoveries, the current one has seen stronger pay increases amongst higher paid professions relative to others (Chart 8). Those with higher levels of relevant education and trainingwhich are unlikely to be those stuck in unemploymenthave more bargaining power than those who are engaged in less remunerative professions. Breaking on Through to the Other Side In an environment where certain in-demand workers are benefitting in ways others cannot, those who Chart 9 Retention Rate Improving For Newly Employed Source: CIBC calculations based on Statistics Canada's tabulations 90 95 100 105 110 115 120 125 93 94 95 96 97 98 99 00 01 02 Low-paying industries Mid-paying industries High-paying industries Wage index Dec 1992=100 3-mos moving avg Not Much Difference During the 1990s Jobless Recovery and the Early 2000 40 45 50 55 60 65 77 81 85 89 93 97 01 05 09 13 Probability of Maintaining Employment Beyond First Year % CIBC WORLD MARKETS INC. Economic Insights - June 18, 2014 6 Source: Statistics Canada, CIBC Chart 11 More Canadians Stay Employed For Longer Share in Total Employment 15 20 25 30 88 90 92 94 96 98 00 02 04 06 08 10 12 14 40 42 44 46 48 50 52 <1 Year job tenure (L) >5 Years job tenure (R) % % Chart 10 Retention Rate Source: CIBC calculations based on Statistics Canada's tabulations do nd employment are increasingly likely to remain employed. Today there is a near-record high 60% chance of remaining employed after completing a rst year on the job 3 (Chart 9), with the retention rate naturally rising with tenure (Chart 10). The share of Canadians with a job tenure of more than ve years is at a record high, and the share of those with less than one year in tenure is at a record low (Chart 11). That marks increased stability in the Canadian labour market. This stable and boring job market is the complete opposite of what was envisioned not too long ago. The job market of the new economy was supposed to permanently alter employer-employee relationships and workers were seen as becoming increasingly disposable, with the implication that job stability would tumble. Rising survival rates between years of employment and increased stability however, makes sense in a world where there is a low supply of newly unemployedand presumably still qualifiedindividuals. The situation today keeps employers motivated to keep workers they have. At the same time, a large overhang of long- term unemployed reduces the motivation of lower skill employees to branch out. The reality for those who are looking for work are less rosy: fewer seats being vacated affords those long-term unemployed with fewer opportunities to nd a way onto payrolls. Couple that with a skills mismatch and an aging population shrinking the workforce, and landing a job becomes harder today than it was previously. But the current environment also suggests that once that higher bar is cleared, a career featuring higher stability lies ahead. To be sure, the business cycle isnt dead and policy still plays a role in rebalancing labour market activity. But the job markets evolving structure makes the balancing act much more challenging. Note: 1) The residual is the difference between the U-6 and U-3 in the US, and the R8 and R3 measures in Canada. The U-3 is the headline unemployment rate in the US. The BLS describes the U-6 as the unemployed included in the U-3, plus all persons marginally attached to the labor force, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all persons marginally attached to the labor force. The R3 is calculated by Statistics Canada in a similar methodology as that of the US headline rate, and the R8 includes under-employed and marginally attached workers that the U-6 also captures. 2) Granted the Beveridge curve presented here has very little history (monthly observations since 2011)but the ndings are consistent with our previous research regarding that issue (see our In Focus titled The Haves and Have Nots of Canadas Labour Market dated December 3, 2012). 3) The retention rate is the probability of an individual with t years of experience remaining employed at year t+1. See Heisz, A. (1996), Change in Job Tenure and Job Stability in Canada, Statistics Canada Catalogue no. 11F0019MPE. 0 20 40 60 80 100 from 1-2 from 2-3 from 3-4 from 4-5 Over 5 Years Likelihood of Maintaining Employment From One Year to Another %