Statistics Canada - Statistique Canada
Skip main navigation menuSkip secondary navigation menuHomeFrançaisContact UsHelpSearch the websiteCanada Site
The DailyCanadian StatisticsCommunity ProfilesProducts and servicesHome
CensusCanadian StatisticsCommunity ProfilesProducts and servicesOther links

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Side menu bar Catalogue Number 75-001-XIE Table of contents Latest issue News from The Daily Latest data Survey information Back issues Feedback Studies Latest issue in PDF

January 2007
Vol. 8, no. 1

Perspectives on Labour and Income

Canada's unemployment mosaic, 2000 to 2006
Ernest B. Akyeampong

The unemployment rate is a well-known barometer of labour-market health. The rise in the national unemployment rate in the years immediately following the high-tech meltdown has been replaced by sustained annual declines, resulting in a rate of 6.3% for 2006. This is not only below the 6.8% registered during the boom, but a 30-year low as well.1

Of course not all parts of the country have shared equally in the improvement. Some have done better, others worse. Normally, comparisons involve the 10 provinces or 5 regions of Canada, but within each, many distinct labour markets can be found. This article focuses on the 28 census metropolitan areas (CMAs) and the 10 provincial non-CMA areas (see Data source and definitions). Using the Labour Force Survey (LFS), the article first tracks unemployment rate dispersion for local labour markets (CMAs and non-CMA areas) between 2000 and 2006. It then examines the comparative labour market performance of these areas based on unemployment rates and rankings, and unemployment duration. Unemployment levels, labour force, and employment are provided in an appendix.

Unemployment rate dispersion rising

The impressive performance of the national unemployment rate in recent years hides considerable geographic disparities. For example, in 2006 the unemployment rate in the Qubec CMA averaged 5.2% compared with 8.4% in nearby Montral. Similarly, the unemployment rate in Kitchener (5.2%) was much lower than in Windsor (9.0%).

That the unemployment rate will differ by geographic area is generally understood. All things being equal, the dispersion is expected to narrow in periods of economic growth, when the national rate is usually falling (Guillemette 2006). However, the reverse has been the case in the current expansion, just as it was in the boom years of the late 1980s (Gower 1996). The variation around the national rate has tended to increase among CMAs and non-CMA areas in the past five years (2002 to 2006) as the national rate has drifted down (Charts A and B) (see Measuring dispersion).

Several reasons have been suggested for the rise in dispersion during the current expansion. First, the economic growth may not be strong or widespread (Guillemette 2006). The current expansion has been strongest in Western Canada (Cross and Bowlby 2006; White, Michalowski and Cross 2006), while performance in some large metropolitan areas such as Toronto and Montral has been more moderate. Others suggest that programs such as Employment Insurance may be discouraging the migration of some unemployed from underperforming areas to 'hot' labour markets, thereby accentuating the dispersion (Guillemette 2006).

Trends and patterns in unemployment rates

Starting from a low of 6.8% in the boom year of 2000, the national unemployment rate rose to 7.2% in 2001, in line with the high-tech meltdown. Unemployment peaked in 2002 (7.7%), stalled the following year at 7.6%, and then declined steadily to 6.3% in 2006 (Chart A). With few exceptions, most areas displayed similar trends (Table 1). The five areas with no clear trends were Prince Edward Island, Windsor, Thunder Bay, non-CMA Ontario, and Regina.

In both 2000 and 2006, Calgary registered among the lowest unemployment rates (4.5% and 3.2% respectively);2 the highest rates were recorded in non-CMA Newfoundland and Labrador (21.3% and 19.3%).

Some areas emerged as perennial best performers, defined here as having the lowest unemployment rates in five of the seven years. Others were perennial poor performers. Nearly all the best performers were in the Prairies (Calgary, non-CMA Alberta, and non-CMA Manitoba, the exception being Victoria). The Alberta areas maintained their enviable position largely as a result of the prosperity brought on by the oil and gas industry and the increased activity in construction. The poor performers were non-CMA Newfoundland and Labrador, Prince Edward Island, non-CMA Nova Scotia, non-CMA New Brunswick, and Windsor.

Both nationally and in a substantial majority of CMAs and non-CMA areas, the unemployment rate in 2006 was lower than in 2000. In eight areas, however, the opposite was true. Except for Montral, the areas were in Ontario, a province hit by reduced activity in manufacturing overall and the auto industry in particular. High energy costs and reduced exports, due in part to the appreciating Canadian dollar, adversely affected these industries. A similar fate befell the manufacturing industries of Montral; particularly hard-hit were its aerospace industry as well as the clothing and textile industry. Montral also saw an employment drop in public administration.

Losses in ranking centred in Ontario

One way of demonstrating the fortunes of the CMAs and non-CMA areas is by way of changes in unemployment rate rank between 2000 and 2006 (Table 2). By this measure, labour markets in Ontario fared worst. Of the 16 areas that saw a deterioration in rank over the period, 9 were in Ontario. In Quebec, Montral and to a lesser degree Sherbrooke also lost some ground, while in Saskatchewan, Regina and the non-CMA areas saw their rankings decline.

Of the five CMAs that registered the largest drops in ranking between 2000 and 2006, four were in Ontario's Golden Horseshoe (Oshawa, Hamilton, Toronto and Windsor) and the fifth was Regina (Table 3). The better performance of the western labour markets is also evident in their strongly positive rank changes. Four of the five areas with the best improvement were in British Columbia: Victoria, Abbotsford, non-CMA British Columbia, and Vancouver. B.C.'s labour market improvements came on the heels of gains in resource-based industries, construction and transportation, and in increased exports to the Far East, notably China. The Qubec CMA also showed a significant improvement in ranking. Industries here registering respectable employment growth included public administration; information, culture and recreation; and transportation and warehousing.

Average unemployment duration falls in most CMAs

Average unemployment duration (weeks of continuous job search) provides one measure of the degree of difficulty faced by those searching for a job (Table 4).3

Unlike trends in the unemployment rate, a positive picture emerges from the average unemployment duration (Chart C). At the national level, duration fell by about 3 weeks (from 19.8 to 16.7 weeks) between 2000 and 2006. Declines were also registered in most areas (33). Whereas 8 areas registered a higher unemployment rate in 2006, only 5 areas had a higher average unemployment duration (Prince Edward Island, Saguenay, Oshawa, non-CMA Alberta, and Victoria). Indeed, except for Oshawa, all areas in Ontario had shorter durations in 2006. The rise in duration in Victoria is intriguing since this CMA was among those registering the best improvement in unemployment rate.

In addition to the fairly steep drop in average unemployment duration in most areas, the degree of dispersion tightened. In 2000, duration ranged from just over 12 weeks in Edmonton and non-CMA Alberta to 33 weeks in Trois Rivires (Table 4). By 2006, it ranged from around 8 weeks in Edmonton to about 22 weeks in Saguenay, Trois Rivires and Montral.


The benefits of the current economic expansion have not been shared equally by the various CMA and non-CMA areas acrossCanada. The unequal distribution is clearly evident in the disparities observed in unemployment rate movements in the different geographical areas.

The past four years have witnessed an improvement in unemployment rates in many areas. Alberta and British Columbia CMAs and non-CMA areas especially have recorded significant improvements, reflecting the boom in oil, gas and other resource-based industries, as well as increased activity in construction and transportation. Only two CMAs, Windsor and Thunder Bay, have seen some recent deterioration or fluctuation in their unemployment rates. In Windsor, this was primarily due to setbacks in manufacturing industries in general and the auto industry in particular. The overall result has been an increase in the unemployment rate dispersion over the past several years.

However, the overall picture emerging from the average duration of unemployment in the 2000s is more encouraging. Not only did the average weeks of continuous job search fall between 2000 and 2006 in most areas, the difference between the shortest and longest also shrank.


Areas in Ontario registered the largest increases in numbers unemployed. This paper examined shifts in unemployment through the unemployment rate and ranking, both measures being abstract. However, the number of people unemployed is also of interest. (Table A1 Unemployment by region)

At the national level, the number of unemployed increased by 2.4% (26,000) between 2000 and 2006. Almost all of the 15 areas registering increases in unemployment numbers were located in Ontario (11) and Quebec (3). The other CMA recording an increase was Regina. Some of the increases were fairly large. For example, unemployment rose in Windsor by 81% (7,000), in Toronto by 38% (54,000), and in Oshawa by 38% (3,000). In Montral, it rose by 19% (27,000). (Table A2 Labour force by region)

The remaining 23 areas recorded decreases in unemployment, with significant declines being registered in Qubec (-28% or -8,000), Edmonton (-22% or -6,000), Victoria (-40% or -4,000), and non-CMA British Columbia (-35% or -24,000). (TableA3 Employment by region)

Data source and definitions

The Labour Force Survey (LFS) collects information each month on labour market activity from the civilian, non-institutionalized population 15 years of age and over. The territories are excluded from the national total, as are persons living on Indian reserves. The survey samples approximately 54,000 households, with each remaining in the sample for six consecutive months.

A census metropolitan area (CMA) consists of an urban core with a population of 100,000 or more, together with adjacent urban or rural areas that have a high degree of economic and social integration with the core. Subtracting CMAs from the provincial total produces residuals consisting of smaller urban and rural areas. These are referred to as non-CMA areas. All of Prince Edward Island is defined as a non-CMA. While these provincial residuals obviously contain many local variations in labour market conditions, such detail is beyond the scope of this article.

The duration of unemployment describes how long (usually in weeks) someone has continuously been looking for a job. The LFS, by design, measures periods of continuous incomplete job search. Information on completed spells can be obtained from longitudinal data sources such as the Survey of Labour and Income Dynamics (SLID).

Measuring dispersion

For a number of reasons, gaps always exist between the national unemployment rate and rates registered by various CMAs and non-CMAs. An increase in the dispersion rate means the gap is widening, and vice versa. In this paper, dispersion rates for CMAs and non-CMA areas are calculated as a weighted mean of the differences between the area and national unemployment rates. Specifically, the absolute difference between each area rate and the national rate is multiplied by the area labour force. These products are summed and the total divided by the national labour force to produce aggregate dispersion. Finally, this is divided by the national unemployment rate to produce percent dispersion.

This is expressed algebraically as:


Ui = unemployment rate in area i

Un = national unemployment rate

LFi = labour force in area i

LFn = national labour force

The dispersion of the average duration of unemployment was calculated in the same fashion.


  1. Caution must be exercised when comparing recent LFS employment and unemployment estimates with those prior to 1976—when the questionnaire underwent significant changes.

  2. In actual fact, in 2000 Calgary's unemployment rate (4.5%) was bettered by that of non-CMA Manitoba (4.3%).

  3. The LFS average durations in Table 4 are, by survey design, for incomplete job search. These are shorter than completed search durations provided by other surveys such as the Survey of Labour and Income Dynamics (SLID). Notwithstanding, the LFS data still provide useful insights on labour market health.


  • Cross, Philip and Geoff Bowlby. 2006. "The Alberta economic juggernaut: The boom on the rose." Canadian Economic Observer. Vol. 19, no. 9. September. Statistics Canada Catalogue no. 11-010-XIB. (accessed January 2, 2007).

  • Gower, Dave. 1996. "Canada's unemployment mosaic in the 1990s - PDF." Perspectives on Labour and Income. Vol. 8, no. 1. Spring. Statistics Canada Catalogue no. 75-001-XPE. p. 16-22. (accessed January 2, 2007).

  • Guillemette, Yvan. 2006. Misplaced Talent: The Rising Dispersion of Unemployment Rates in Canada. Toronto. C.D. Howe Institute. 5 p.

  • White, P., M. Michalowski and P. Cross. 2006. "The west coast boom." Canadian Economic Observer. Vol. 19, no. 5. May. Statistics Canada Catalogue no. 11-010-XIB. (accessed January 2, 2007.

Full article in PDF

Ernest B. Akyeampong is with the Labour and Household Surveys Analysis Division. He can be reached at 613-951-4624 or

You need to use the free Adobe Reader to view PDF documents. To view (open) these files, simply click on the link. To download (save) them, right-click on the link. Note that if you are using Internet Explorer or AOL, PDF documents sometimes do not open properly. See Troubleshooting PDFs. PDF documents may not be accessible by some devices. For more information, visit the Adobe website or contact us for assistance.

Home | Search | Contact Us | Français Return to top of page
Date Modified: 2014-05-14 Important Notices