Posted: Thursday, 6 March 2014
A Profile of Canada's Labour Market
Each month, Statistics Canada releases its unemployment rate, and many use it to talk about job creation in Canada. Monthly unemployment rates move up and down, making headlines but revealing little. As work patterns change, with greater use of part-time employees and other forms of precarious labour, the headline unemployment rate becomes less and less useful on its own. The labour force is comprised of far more than simply employed and unemployed workers.
A broader and longer-term analysis, and better labour market indicators are required to give insight into the various ways that workers in Canada responded to the recession and weak economic recovery. This paper takes an in-depth look at the recovery and the current state of the labour market, to highlight current challenges in the way we look at the labour market, and propose alternative indicators that should be part of Statistics Canada’s monthly releases to better inform the public about the real state of the labour market.
Slowing Job Growth
Canada's labour market recovered the total number of jobs lost during the recession by November 2010, leading to claims of a robust recovery for workers. A closer look shows that job creation has stagnated and hasn't kept up with population growth among working age adults.
When we set up an index job level at 100 prior to each of the last three recessions—as shown in Graph 1—we find that although the most recent recession had the most rapid loss of jobs, employment losses were relatively shallow and employment returned to pre-recession levels earlier than prior recessions. Nearly five years out, however, it's a different story. Growth in employment now lies somewhere between the previous two recessions, mirroring the long, slow recovery path following the 1990's recession.
The employment rate, that is the proportion of the working age (15-64) population employed shown in the Graph 2, has recovered only half-way to its pre-recession peak. This indicator also shows employment growth has completely stalled over the past year and half.
The jobs that have been recovered are disproportionately part-time and precarious. As seen in Table 1, part-time jobs grew at twice the rate of full-time jobs (5.9% vs 3.3%), and account for 40% of the job growth between 2008 and 2013—even though part-time positions only make up one out of five jobs (19%) in the labour market. All of the growth in part time jobs was among underemployed part time workers, meaning those who want more hours of work.
Unemployment rate has stalled and does not tell us the full story
Nearly five years after the end of the 2008-2009 recession, Canada's headline unemployment rate has remained fixed at 7.2% (December 2013), a level first reached mid-2011. But rather than a static group of individuals, large numbers of workers flow in and out of unemployment each month. Recessions affect these flows in various ways, for example fewer people quit their job or decide to enter the labour market when the job market looks dismal. Following a recession, there is usually an increase in the number of workers that are unemployed for extended periods of time.
The proportion of workers who were unemployed for over 12 months rose dramatically following the prolonged 1990—1992 recession (Graph 3), accounting for over 15% of unemployed workers five years after the recession was officially over.
The rise in long term unemployment was not as dramatic following the 2008-2009 recession, but remains a concern at 12% of all unemployed workers, double the pre-recession level.
Underemployment is larger than Unemployment
Underemployment can be thought of as the unmet need for paid employment. For example, a person may be unable to find full-time work, but manages to find part-time work. There is still an unmet need for more hours of work, even though this person is no longer counted as unemployed. A person may want a job, and be actively seeking work, but they are not immediately available. Alternatively, a person may wish to work, but has given up searching, and possibly even engaged in unpaid activities such as care work. In each of these cases, usually grouped together as marginal labour force attachment, there is still an unmet need for paid employment.
Statistics measuring these concepts are used by Statistics Canada to generate supplementary unemployment rates, labelled R5 through R8. Unfortunately, these supplementary unemployment rates are more restrictive than the current international standard1. This affects who gets counted as underemployed.
Statistics Canada only counts the volume of underemployment, meaning the total hours of underemployment divided by average weekly full-time hours.1 This avoids measuring the actual number of people affected by this phenomena. The international consensus on reporting time-related underemployment is the count of individuals who fall into this category. Graph 5 looks at both the volume and the total count of time-related underemployment.
The portion of underemployment counted by Statistics Canada has been consistently around half the amount of the total persons that are underemployed part-timers since 1997. In 2013, the total number of employed persons seeking more hours was over 910,000, and the amount counted by Statistics Canada was only 445,000. That considerably alters the level of underemployment reported by Statistics Canada.
Even more worrying, the amount of underemployment that is not captured by Statistics Canada’s measure varies greatly by group. For example, underemployment among women is significantly underestimated, because women are far more likely to be part-time workers who want more hours.
At the Margins of the Labour Force
In the Labour Force Survey, Statistics Canada collects information on those who are not in the labour force but wanted work. Unfortunately, Statistics Canada only includes a subset of these persons in their estimates of underemployment, those who weren't looking because they feel there is no work available (discouraged workers), and those who are waiting for recall or reply from an employer.
Since the Labour Force Survey started collecting this information, there has been a significant shift in the reasons why workers aren't looking for work, even though they would like a job (Graph 6).
In 1997, discouraged workers made up nearly 25% of all workers who indicated that they would like a job but weren't looking. Since 2006, this had fallen to just over 5%. Those awaiting recall or reply, the other element that Statistics Canada includes in its supplementary unemployment measure, has followed much the same trend.
On the other hand, the proportion of individuals who are at school but want a job has increased from 20% of all responses to nearly 30% over the past 15 years. The catch-all category of “other”, has increased from 15% to just over 25%. Why not count them as underemployed?
The result of the changing make-up of responses to this question means that the two reasons that are included in Statistics Canada’s estimates of underemployment have fallen from 40% of all reasons for not looking, to only 12%. This means that the overwhelming majority of marginally attached workers are not counted in any of Statistics Canada's supplementary unemployment rates. The impact of this is most noticeable among young workers 15-24.
Broader measurement of underemployed is needed. If we include all persons with some attachment to the labour force, and all involuntary part-time workers, we see a marked difference in underemployment rates, especially for young workers and women (Graph 7).
Comparing the change in labour statistics for 2008 and 2013 tells us that the Canadian labour market has yet to return to a healthy pre-recession position in terms of the number of jobs, or the types of jobs.
The economy has not created enough jobs, and those that have been created are disproportionately precarious. 72% of net new jobs created between 2008 and 2013 fall into the precarious or underemployed categories. More Canadians are unemployed, marginally attached, or simply not engaging in the labour force.
Since 2011 the number of underemployed workers has exceeded the number of unemployed workers – in 2013 there were 1.35 million unemployed workers and 1.43 million additional underemployed workers. And that is before we even begin to take into account skills-related underemployment. This is an issue that needs to be taken seriously.
Before we can address an issue, we first must understand it. A broader view of labour market information gives us a better picture of who is underemployed. It is useful to think about the labour force as more dynamic than a single indicator can measure, and more complex than simply employed and unemployed workers. Many categories overlap, and often individual workers move between categories from month to month.
This diagram helps us think about the various ways that workers can be employed or underemployed. For example, a part-time worker who is looking for full-time work is both employed and underemployed. Those who would like a job but aren't actively looking are both underemployed, and not in the labour force.
Looking at the labour market this way gives us a better start on how we should develop policy. For example, supplementary unemployment statistics tell us that underemployment is a huge and hidden issue for women. The underemployment rate for young workers underlines the serious barriers too many face in entering the labour market. The breakdown of labour market statistics tells us that the quality of jobs is not where we would like them to be. Finally, we recognize that this is an issue which affects many more people than we had thought - the underemployment rate for 2013 was 14.2%, double the headline unemployment rate of 7.1%.
A good, and quite feasible, first step would be to publish the broader measure of underemployment that is described in this paper—but we need to go further.
Canada needs to develop an employment strategy that supports the creation of better paid and more secure jobs for everyone. Workers in Canada need action now to build a prosperous economy that everyone can share in.