Insights on Canadian Society
Linking labour demand and labour supply: Job vacancies and the unemployed

Introduction

Observers of the Canadian labour market are familiar with standard measures of the labour supply and labour demand for workers. The unemployment rate provides information on the supply of persons looking for work in excess of those who are currently employed. Data on employment provide information on the demand for workers that is already met by employers. Additional information about the demand side of the labour market is collected by the Job Vacancy and Wage Survey (JVWS). As of January 2015, the JVWS provides an indicator of unmet labour demand by asking employers to report the number of job vacancies at their business location.^Note 1

By linking unmet labour demand (job vacancies) and excess labour supply (unemployed persons), a number of labour market dynamics may be observed, particularly how easily and efficiently workers find jobs and whether the skills employers are looking for differ from the ones available in the labour market. The availability of unfilled jobs—the number of job vacancies or the job vacancy rate—and its relationship to unemployment is an important measure of how tight or slack the labour market is.

This paper explores the additional analytical insights that can be gained from the joint availability of job vacancy and unemployment data.^Note 2 This study uses data from the Job Vacancy and Wage Survey (JVWS) and Labour Force Survey (LFS) to answer the following questions: (1) To what extent are job vacancies and unemployment related? (2) What can the unemployment-to-job vacancy ratio tell us? (3) To what extent do occupations differ in their relative degree of being "slack" or "tight"?

Three dimensions of labour demand and supply are examined in this article: differences across provinces, differences across occupations, and differences across levels of education. Prior to discussing the results, the next section will examine some conceptual issues related to unemployment and job vacancies.

Job vacancies and unemployment

The definition of job vacancies and unemployment are conceptually similar. The three conditions for a job to be considered vacant mirror the three conditions for a person to be considered unemployed. For a job to be considered vacant, a position must be currently available, work could start within 30 days and an employer must be actively recruiting someone to fill the job from outside the organization.^Note 3 To be considered unemployed, a person must be available for work, could start work immediately and must be actively searching for work.

Job vacancies and unemployment coexist in a dynamic labour market.^Note 4 In some situations, vacancies and unemployment are associated with normal turnover in the labour force. Individuals join or return to the labour market to search for work, while others leave the labour market to retire, go to school or work in the home. Similarly, new jobs are created while other jobs disappear. Employers and job seekers need time to be brought together. In other situations, workers lack the skills to meet the job requirements or live too far from regions where jobs are available. Finally, some firms or industries may respond to fluctuations in economic activity with temporary layoffs. While those on temporary layoff are considered unemployed, job vacancies do not necessarily exist since the unemployed are waiting for their previous job to reappear.^Note 5

In the two-year period from January 2015 to December 2016, there was an average of about 1.3 million unemployed persons in Canada and about 390,100 vacant jobs. That said, in any given quarter, the number of unemployed persons is greater than the number of vacant positions (Table 1).

Job openings move in the opposite direction of unemployment over the course of the business cycle. This theory is supported by evidence from the United States, which suggests that job vacancies fluctuate in a countercyclical movement with the number of unemployed persons.^Note 6

During economic expansions, the demand for goods and services strengthens. The labour market is described as tight with employers searching for employees: most people who want a job are already working, unemployment is low and job vacancies tend to be high. As the economy contracts, the demand for goods and services weakens. The labour market is described as slack with employers cutting back on recruiting new employees, while the few job vacancies that do exist tend to be filled quickly. Unemployment is higher due to a reduction in overall hiring relative to a decline in workers quitting and to an increase in layoffs.

With such a short times series, it is not possible to formally evaluate how Canadian job vacancies and unemployment jointly vary over time.^Note 7 For this reason, this article focuses on the estimates of job vacancies and unemployment averaged over the eight quarters of data from January 2015 to December 2016.^Note 8 The Data sources, methods and definitions section provides a detailed description of how the data are used in the current context.

The relationship between the job vacancy rate and the unemployment rate can be shown in a chart (Chart 1). Each data point corresponds to the average job vacancy rate and the average unemployment rate in an economic region over the eight quarters of data from 2015 to 2016. There are a total of 66 economic regions.^Note 9

Data table for Chart 1

Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1. The information is grouped by Unemployment rate (percent) (appearing as row headers), Job vacancy rate (percent) (appearing as column headers).

Unemployment rate (percent)	Job vacancy rate (percent)
10.1	1.9
18.8	2.1
17.3	2.2
11.1	2.3
15.0	1.7
9.4	2.2
8.3	2.6
11.2	2.4
6.4	2.6
16.3	1.9
8.9	2.6
8.6	2.5
8.7	1.9
8.2	2.3
15.4	1.6
8.2	1.6
5.0	1.8
5.4	1.9
6.4	1.6
6.1	1.7
6.1	1.8
10.1	2.0
7.9	2.0
7.6	1.5
6.0	1.7
8.8	1.7
6.9	1.6
7.6	1.5
8.3	1.2
9.5	1.7
7.3	2.4
6.8	2.3
6.9	2.5
7.1	2.8
5.5	2.8
6.3	2.6
6.1	2.4
7.6	2.5
5.2	2.6
7.7	2.3
6.6	2.3
5.1	1.8
4.2	1.9
5.0	2.3
6.6	2.1
5.9	2.2
5.4	3.0
4.9	2.1
3.9	3.4
6.0	2.3
5.6	2.5
8.0	2.4
6.1	2.3
6.3	2.5
7.9	2.5
6.5	3.6
7.5	2.3
6.9	2.8
8.8	2.5
6.3	2.7
5.9	3.4
7.3	3.7
7.9	3.1
7.4	2.8
7.4	2.8
7.8	3.3
Sources: Statistics Canada, Job Vacancy and Wage Survey; Labour Force Survey, 2015 and 2016.

There is a modest negative yet statistically significant relationship between unemployment and job vacancies: a high rate of unemployment is typically associated with a low job vacancy rate and, conversely, a high job vacancy rate coincides with a low unemployment rate. To test the robustness of the relationship between the vacancy rate and the unemployment rate, a linear regression was performed. The results indicate that a 1.0 percentage point increase in the unemployment rate is associated with a 0.9 percentage point decline in the average job vacancy rate.^Note 10 It should be noted that, at most, 9% of the cross-regional variation in job vacancy rates is explained by the unemployment rate in the economic region.

An alternative chart can be produced by sorting the economic regions from eastern Canada to western Canada and plotting both the job vacancy rate and unemployment rate (Chart 2). In economic regions with high job vacancy rates (as seen in the western provinces), the unemployment rates are typically lower, while economic regions with low job vacancy rates typically have higher unemployment rates (as seen in the Atlantic provinces and Quebec).^Note 11

Data table for Chart 2

Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2. The information is grouped by Economic region (appearing as row headers), Unemployment rate and Job vacancy rate, calculated using Unemployment rate (percent) units of measure (appearing as column headers).

Economic region	Unemployment rate	Job vacancy rate
	percent
Atlantic
1010	10.1	1.9
1020	18.8	2.1
1030	17.3	2.2
1110	11.1	2.3
1210	15.0	1.7
1220	9.4	2.2
1230	8.3	2.6
1240	11.2	2.4
1250	6.4	2.6
1310	16.3	1.9
1320	8.9	2.6
1330	8.6	2.5
1340	8.7	1.9
1350	8.2	2.3
Quebec
2410	15.4	1.6
2415	8.2	1.6
2420	5.0	1.8
2425	5.4	1.9
2430	6.4	1.6
2433	6.1	1.7
2435	6.1	1.8
2440	10.1	2.0
2445	7.9	2.0
2450	7.6	1.5
2455	6.0	1.7
2460	8.8	1.7
2465	6.9	1.6
2470	7.6	1.5
2475	8.3	1.2
2480	9.5	1.7
Ontario
3510	7.3	2.4
3515	6.8	2.3
3520	6.9	2.5
3530	7.1	2.8
3540	5.5	2.8
3550	6.3	2.6
3560	6.1	2.4
3570	7.6	2.5
3580	5.2	2.6
3590	7.7	2.3
3595	6.6	2.3
Prairies
4610	5.1	1.8
4620	4.2	1.9
4630	5.0	2.3
4650	6.6	2.1
4660	5.9	2.2
4670	5.4	3.0
4710	4.9	2.1
4720	3.9	3.4
4730	6.0	2.3
4740	5.6	2.5
4750	8.0	2.4
4810	6.1	2.3
4820	6.3	2.5
4830	7.9	2.5
4840	6.5	3.6
4850	7.5	2.3
4860	6.9	2.8
4880	8.8	2.5
British Columbia
5910	6.3	2.7
5920	5.9	3.4
5930	7.3	3.7
5940	7.9	3.1
5950	7.4	2.8
5960	7.4	2.8
5980	7.8	3.3
Sources: Statistics Canada, Job Vacancy and Wage Survey; Labour Force Survey, 2015 and 2016.

In 2015/2016 there were 3.4 unemployed persons per job vacancy in Canada

This section introduces the ratio of unemployed persons per job vacancy. The unemployment-to-job-vacancy ratio combines both labour supply and demand measures by gauging excess supply of labour relative to the unmet needs of Canadian employers. This ratio is constructed by using the number of unemployed persons and dividing it by the number of job vacancies. The number of unemployed persons is a measure of persons looking for work in excess of those who are currently employed. Job vacancies are a measure of the demand for workers not currently met by employers. While there are some caveats when using JVWS and LFS data sources together,^Note 12 this ratio is a simple way of examining the composition of labour supply as it relates to labour demand.

The ratio is often used to describe how tight or slack the labour market is. Lower values of the ratio imply that there are fewer unemployed persons per job vacancy and possibly greater ease of finding a new job, suggesting a tight labour market. Conversely, higher values of the ratio imply that there are more unemployed persons per job vacancy and possibly greater difficulty finding a new job, suggesting a slack labour market.

The ratio of unemployed persons per job vacancy is sensitive to the measure of unemployed persons used and raises the following conceptual issues:

1. The Labour Force Survey classifies unemployed persons into the three following groups: those who are searching for a job, those on temporary layoff, and those who are starting a new job in the next four weeks. Using this measure results in 3.4 unemployed persons per job vacancy.
2. While it is clear that job searchers should be included in the ratio, it is not clear whether those on temporary layoff and future job starters should be included. Persons on temporary layoff are expecting to be recalled to their previous position, while future job starters include those who are starting a new job within the next four weeks. When attention is restricted to unemployed job searchers, the result is 3.2 unemployed job searchers per job vacancy.
3. Hidden unemployment refers to situations in which individuals may be without work but are not classified as unemployed according to official statistics. Such individuals include discouraged workers or individuals who are not actively seeking work because they believe that no work is available but wish to work at the prevailing wage rate. In addition, some workers find themselves underemployed. An example is part-time workers actively searching for full-time work. When discouraged workers and underemployed individuals are counted as unemployed, the result is 4.2 persons who want to work per job vacancy.

Each of these examples illustrates the conceptual difficulties in accounting for the unemployed in the ratio of unemployed persons to job vacancies. For the remainder of this article, unemployed refers to those who are searching for a job, on temporary layoff and starting a new job in the next four weeks, and is used in the ratio of unemployed to job vacancies. This is consistent with the literature and other international evidence.^Note 13

In the next section, the labour market is partitioned into characteristics such as geography, occupation and education. Showing how labour market dynamics vary along these dimensions yields further insight into how the labour market functions.

British Columbia has the lowest unemployment-to-job vacancy ratio whereas Newfoundland and Labrador has the highest

There are provincial differences between those searching for work and where job opportunities exist (Table 2). A larger ratio of unemployed persons per job vacancy can occur when there is a larger share of unemployed persons relative to their share of job openings. For example, Quebec has a higher share of unemployment (24%) relative to its share of job vacancies (16%). This is also the case in Newfoundland, which has 3% of the unemployed, but 1% of job vacancies. The slack labour market conditions in these provinces are reflected in the high ratios of unemployed persons per job vacancy: there were 5.2 and 8.2 unemployed persons per job vacancy in Quebec and Newfoundland, respectively.

Ontario’s share of unemployed persons (37%) is relatively close to its share of job openings (41%). There are 3.1 unemployed persons per job vacancy in Ontario, a rate lower than in Quebec and the Atlantic provinces.

A small ratio of unemployed persons per job vacancy can occur when a province has a lower share of unemployment relative to job openings. British Columbia has a relatively lower share of unemployment (11%) relative to its job vacancies (18%). The tighter labour market conditions in British Columbia are reflected in the low ratio of unemployed persons per opening (2.2 unemployed persons per job vacancy).^Note 14

Worker mobility is traditionally from areas of high unemployment towards areas where job opportunities are more abundant.^Note 15 The federal government publication Jobs Report: The State of the Canadian Labour Market suggests that only 1% of Canadians move across provincial borders and 3.4% move across census subdivisions within a province in a given year. The decision of unemployed workers to migrate to regions with higher job vacancies is often complicated and may be influenced by a number of factors such as age, income, marital status, presence of children, language, culture, distance from home province and position in the business cycle, as well as institutional factors such as credential recognition.^Note 16

Provincial differences in unemployment and job vacancies may reflect a variety of other differences, most notably those by occupation and industry.

There are more job vacancies than unemployed persons in health occupations in all provinces

This section reports differences in the ratio of the unemployed to job vacancies by occupation group. Occupations are grouped according to the work usually performed as determined by the tasks, duties and responsibilities of the occupation.^Note 17

The LFS collects information on the occupation of the previous job held during the past year for individuals who did not have a job during the reference week. Currently unemployed persons who held a job in the past year represent about 58% of the total unemployed population. As such, they may not represent the experience of all unemployed persons in the Canadian labour market. In addition, these unemployed persons may find it easier to secure future employment given that they have recent and relevant work experience in the occupation.^Note 18

Since currently unemployed persons who held a job in the past year are a subset of total unemployment and holding the number of job vacancies constant, the ratio of unemployed persons per job vacancy is smaller than previously reported. Overall, there are 2.0 unemployed persons who held a job during the past year per job vacancy (compared with 3.4 unemployed persons per job vacancy). The ratios of unemployed persons who held a job in the past year to job vacancies are shown for each major occupation group in Chart 3.

Data table for Chart 3

Number of unemployed persons^{Chart 3 Note 1} for each job vacancy, by major occupation group, 2015 and 2016
Table summary
This table displays the results of Data table for Chart 3. The information is grouped by Occupation group (appearing as row headers), Ratio (appearing as column headers).

Occupation group	Ratio
Health	0.7
Management	1.3
Sales and service	1.6
Natural and applied sciences and related	1.8
All occupations	2.0
Business, finance and administration	2.1
Education, law and social, community and government services	2.5
Art, culture, recreation and sport	2.6
Manufacturing and utilities	2.9
Natural resources, agriculture and related production	3.0
Trades, transport and equipment operators and related	3.5
Note 1 The Labour Force Survey collects information about the occupation of the previous job held during the past year for individuals who did not have a job during the reference week. Return to note 1 referrer Sources: Statistics Canada, Job Vacancy and Wage Survey; Labour Force Survey, 2015 and 2016.

Workers in health occupations face tight labour market conditions. Of all major occupation groups, health occupations report the lowest unemployment-to-job vacancy ratio. The ratio (less than 1.0) indicates that there are more job vacancies than unemployed workers in health occupations: for every job vacancy reported for health occupations, there are 0.7 unemployed health workers. Within the health field, detailed occupations such as professional occupations in nursing, technical occupations in health, and assisting occupations in support of health services have ratios of less than 1.0 (Table 3).^Note 19

There are 1.6 unemployed persons per job vacancy in sales and service occupations. This broad occupation group accounts for a disproportionately high number of job vacancies (37%) and a high number of unemployed persons (29%). This suggests that employers can easily find the workers they need and that employees can easily transition from job to job in these occupations—which is the case in the retail and accommodation and food industries, where sales and service occupations are predominately found.^Note 20

Trades, transport and equipment operator occupations are characterized by high unemployment and low job vacancies, resulting in a relatively high ratio of unemployed persons per job vacancy. About 202,830 unemployed persons worked in trades, transport and equipment operator occupations in the previous 12 months—with 57,510 job vacancies in those occupations—resulting in 3.5 unemployed persons per job vacancy. The data reported for the construction and manufacturing industries corroborate these findings, and may suggest there is currently a surplus of workers in such industries.^Note 21

There is heterogeneity within major occupation classification groups. Generally speaking, professional occupations have lower ratios of unemployed persons per job vacancy than technical or support occupations within the same major occupation groups. For example, there are 1.4 unemployed persons per job vacancy among natural and applied science professionals, compared with 2.2 unemployed persons per job vacancy among technical operations related to natural and applied sciences. Similar conclusions are found in the broad business and finance occupation group.^Note 22

It is important to note, however, that the ratio of unemployed persons per job vacancy may not be always indicative of a slack versus tight labour market (relative to an all-occupation average or a national average). In some cases, a low ratio may also simply reflect an occupation characterized by a high degree of labour turnover. For instance, high turnover occupations (which are predominantly low skilled) may have a large number of vacancies relative to the number of unemployed persons (hence a low ratio). These jobs are not in high demand and the job market is not tighter in these lower-skilled jobs than in higher-skilled jobs; rather, this situation is due to worker turnover and the resulting job vacancies are high for these positions.^Note 23 Examples of such occupations may include sales occupations and personal and customer information services, which are characterized by a relatively high number of job vacancies (more than 40% of the total), and a relatively low ratio of unemployed persons to vacancies.

The previous discussion showed how the ratio of the unemployed per job vacancy varied by geography and occupation independent of one another. This section combines this information in order to gain a better understanding of the labour market situation they represent.

In this scenario, the assumption is that the unemployed are searching in the same occupation and region where they worked in the previous 12 months. While some job applicants may not be searching for new opportunities in their previous field, their recent and relevant work experience in their last occupation may help them be more successful in their job search.^Note 24 Some provinces are grouped to allow for adequate sample sizes for analysis.

On one hand, workers in health occupations face the tightest labour market conditions in all regions of the country. The ratio is less than 1.0 regardless of whether the region is characterized by a tight (as in the case of British Columbia) or slack (as in the case of the Atlantic provinces) labour market (Table 4).

On the other hand, signals are mixed for occupations in natural resources, agriculture and related production. The labour market for this occupation is described as slack in the Atlantic provinces (similar to slack regional labour market conditions) where there are 13.3 unemployed persons per job vacancy, and as tighter in Ontario where there are 1.7 unemployed persons per job vacancy.

It is possible that there are more job opportunities in a province in occupations other than in the occupation in which the individual is searching. It may be the case that job seekers need to retrain to get job offers in different occupations, but they may not be willing to retrain given costly investments in human capital. Others may hold out for a position requiring their skills and may not seek or accept job opportunities in alternative occupations that may offer lower wages than they earned in the past.

About two-thirds of job vacancies require no more than a high school education, but nearly one-half (49%) of the unemployed have a postsecondary education

The ratio of unemployed persons to job vacancies falls as education levels increase. Over one-third of all job vacancies require no minimum level of education. These jobs are available to all unemployed persons, regardless of their education level. As such, there are potentially 8.7 unemployed persons competing for each job vacancy that requires no minimum level of education (Chart 4).

Data table for Chart 4

Data table for Chart 4
Table summary
This table displays the results of Data table for Chart 4. The information is grouped by Education level required (appearing as row headers), Ratio (appearing as column headers).

Education level required	Ratio
University degree and above	0.7
College and below university	1.6
All unemployed persons	3.4
High school graduate	3.8
No minimum level of education required	8.7
Note: Job vacancies that require an apprentice or trade certificate or diploma are excluded. Also excluded are unemployed persons with that level of education. Sources: Statistics Canada, Job Vacancy and Wage Survey; Labour Force Survey, 2015 and 2016.

As the education requirement for job openings increases, the potential pool of applicants with at least the minimum level of required education gets smaller while the number of available jobs gets larger. Unemployed persons with a university education are qualified for vacant jobs requiring a university education and for all job vacancies requiring less education. For this group, there are more job vacancies available than there are unemployed persons. This translates into a ratio of less than 1.0.^Note 25

While university-educated persons benefit from having the most jobs available to them, in most cases their educational qualifications exceed those required for the vacant positions. In other words, they may be overeducated or overqualified for 9 out of 10 available jobs. They may not accept positions requiring lower skill levels and may prefer to hold out for a position more closely related to their skills.^Note 26 Similarly, people with a trades certificate or diploma may be looking for specific job openings that correspond to their qualification level. It is thus important to examine how many job openings correspond to unemployed persons in every category of educational attainment.

If unemployed university-educated persons restricted their job search only to jobs requiring a university education (or to jobs that they are adequately qualified for), there would be over 5.9 unemployed persons per job opening (Table 5).^Note 27 Conversely, restricting vacancies that require a high school education only to those with this level of education would result in 3.8 unemployed persons per vacancy.

About two-thirds of job vacancies require no more than a high school education, but nearly one-half (49%) of the unemployed have a postsecondary education. These results reflect the fact that job vacancies are more likely to demand lower skills than those possessed by the unemployed population. Again, this may be partially attributable to high worker turnover and the relatively large number of job vacancies in low-skilled jobs. It may also be that jobs requiring a higher level of education are more likely to be advertised internally within firms, or that these jobs do not necessarily meet the requirement that work must begin within 30 days.

The unemployed may not be filling available jobs

Not all persons filling jobs come from the pool of unemployed persons. Each year, thousands of workers in Canada start a job with a new employer. Some are new entrants into the labour force or re-entrants searching for work, while others move quickly into new jobs without an intervening job search period.

Previous Canadian research has shown that worker turnover is a salient feature of the Canadian labour force. Regardless of general economic conditions, a large volume of worker reallocation across employers takes place each year.^Note 28 Several indicators are used to show the importance of such reallocation in the current context of job vacancies and unemployment. First, according to the Labour Force Survey, about 19% of the unemployed in any given month move into a paid job the next month.^Note 29 Second, LFS data show that about 6.5% of Canadian workers started a new job in the last three months. Third, according to the Longitudinal and International Study of Adults, 13% of employed Canadian workers looked for a new job in 2014.^Note 30

In general, more workers started a new job in the last three months than the number of reported job vacancies, which may be the result of how vacancies are reported. Some jobs may become vacant and filled during the month. For other jobs, recruiting and hiring may take place long before the job start date (i.e., a period longer than the condition used to define job vacancy that work must begin within 30 days). Pools of successful candidates may be created to fill current and/or future vacancies, as may be the case in larger firms with more formal and lengthy hiring processes, or in industries or occupations that require specialized skills. Both of these very short-term and longer-term vacancies may not be included in this paper’s measure of job vacancies.

Conclusion

This article provides a simple framework to compare job vacancies and unemployment. It attempts to explore the additional analytical insights that can be gained from the joint availability of job vacancy and unemployment data.

This paper began by showing that job vacancies are negatively correlated with the unemployment rate. As more data become available, analysts may be able to distinguish between cyclical and frictional or structural explanations in both level and trend. Job vacancies may become a leading indicator of changes in the business cycle as changes in job vacancies may predate changes in the unemployment rate.

This analysis has also shown how job vacancies and unemployment co-exist at the provincial, occupation and education levels. This adds to the understanding of labour market conditions described as tight or slack. For example, health occupations have a relatively high number of vacancies relative to the number of unemployed persons with recent work experience in health occupations. Conversely, the labour market has a relatively high number of unemployed persons who work as trades, transport and equipment operators relative to the number of job vacancies in those occupations. This result points to the difficulty matching jobs and workers.

Lastly, the results have shown that there are discrepancies between the characteristics of the vacancies and the characteristics of the unemployed. Job vacancies are largely concentrated in sales and service occupations, while the unemployed are proportionately less likely to have been employed in those occupations. Furthermore, two-thirds of job vacancies require no more than a high school education, while nearly one-half of the unemployed have a postsecondary education.

Readers should note, however, that not all job vacancies are filled by the unemployed. Job vacancies are sometimes filled by discouraged workers, underemployed workers, and even employed workers who are looking for a new job.

Marie Drolet is a senior analyst with the Labour Statistics Division at Statistics Canada.