Local data alert! In a series of articles in March, I will analyze local statistics on the number of workers in arts, culture, heritage, and (some) sport occupations in 31 Census Metropolitan Areas (CMAs), using data from 2013 to 2025. The data are available for CMAs, not for municipalities per se.
Today, I’ll highlight the trends for Canada, the 31 CMAs (as a group), and the 6 largest CMAs (individually and as a group). Each of the 6 largest CMAs has a population of over 1.5 million: Calgary, Edmonton, Montreal, Ottawa-Gatineau, Toronto, and Vancouver.
In upcoming articles, I’ll provide a similar analysis for 25 other CMAs: Abbotsford-Mission, Barrie, Fredericton, Greater Sudbury, Guelph, Halifax, Hamilton, Kelowna, Kingston, Kitchener-Cambridge-Waterloo, London, Moncton, Oshawa, Peterborough, Québec, Regina, Saint John, Saskatoon, Sherbrooke, St. Catharines-Niagara, St. John’s, Trois-Rivières, Victoria, Windsor, and Winnipeg.
I’ll analyze these CMAs in three different articles: 1) the 7 CMAs with populations between 500,000 and 1 million; 2) the 10 CMAs with populations between 220,000 and 500,000; and 3) the 8 CMAs with populations under 200,000.
The articles are based on publicly available occupation statistics from the Labour Force Survey (LFS). The LFS has a sample size of 56,000 households each month. The data are imperfect but useful, I think. I should qualify my understanding of their “usefulness” by saying that data on smaller population groups have higher margins of error.
For today’s article, I have created three-year averages to provide a higher level of statistical reliability and minimize the impact of unusual single year estimates in many CMAs. The data in this article labelled 2025 represent the average for 2023-2025. The earliest data point (labelled 2013) is actually the average for 2011-2013. This technique is similar to what I did in my recent analysis of multiple job holding by province.
Focus on broad, longer-term trends
Given the Labour Force Survey’s limitations (which I outline at the end of this post, and in even greater detail here), I will stick to some broad trends among the grouping of workers in the arts, culture, heritage, and sports.
The analysis that follows is relatively simple: I examine whether the number of workers in the selected occupation grouping has increased or decreased over a relatively long timeframe.
Why use the Labour Force Survey, given its limitations?
I’ve analyzed this dataset (now), rather than the census (which has a much larger sample size) because of timeliness and easy availability.
The strength of the Labour Force Survey is its frequency. It is conducted monthly, but the local cultural estimates would not be reliable on a monthly basis. However, when multiplied by 12 for each year, then by 3 for the three-year averages, the sample sizes become more interesting.
The annual averages from the Labour Force Survey can show us trends that the census, conducted only every five years, cannot. In addition, occupation data from the census is published roughly a year and a half after the census is conducted. We will probably be waiting until late 2027 for labour force data from the census in May of 2026.
Occupational categories
Because of its relatively small sample size, the Labour Force Survey provides reliable data only for a broad category of arts, culture, heritage, and sports workers. The summary data relate to what Statistics Canada calls “occupations in the arts, culture, recreation, and sports, except management”.
Most of the people in this category are cultural workers: I have estimated that roughly 75% to 80% of people in this broad occupation grouping work in the arts, culture, and heritage. And the category covers most (but certainly not all) cultural workers: I have estimated that the workers in this broad occupation grouping account for roughly 60% of all workers in the arts, culture, and heritage. (Further details, including a thorough discussion of LFS limitations, is available here.)
Within this broad occupational category, there are four occupation groupings. In Statistics Canada’s language, these are:
Professional occupations in the arts and culture, which include: some artists, such as producers, directors, conductors, and musicians; writers, translators, and other communications professionals; as well as librarians, archivists, conservators, and curators.
Technical occupations in the arts and culture, including: graphic and interior designers; as well as technical workers in libraries, archives, motion pictures, broadcasting, and the performing arts.
Other occupations in the arts, culture, and sports, including: artists such as dancers, actors, comedians, circus performers, photographers, craftspeople, and visual artists; theatre and fashion designers; select museum and art gallery jobs (e.g., registrars, restorers); assistants in film, broadcasting, photography, and performing arts; as well as a few sports occupations (athletes, coaches, and referees).
Support occupations in the arts, culture, and sports, including: program leaders and instructors in recreation, sport, and fitness; as well as puppeteers, buskers, magicians, portrayers of Santa Claus, influencers, and models. The recreation and sports workers in this occupational grouping greatly outnumber the cultural workers (which is why I exclude this grouping whenever possible).
Earnings data are not available in the summary dataset published by Statistics Canada. For the cultural sector, this is not a big loss, because the LFS questionnaire captures salaries only, not self-employment earnings. (That’s the more important limitation for the cultural sector and for other sectors where self-employment is common.)
Canada-wide data
The first graph shows the nationwide data for each three-year period. The vertical axis is in the thousands. I will focus on the trends over time, more than the raw numbers, mostly due to the occupational category that is imperfect for our purposes. (As noted above, cultural workers represent about three-quarters of all workers in the category, which includes roughly six in every ten cultural workers.)
There has been a relatively consistent increase in the number of workers in the arts, culture, heritage, and sport since 2013. The exception was around the pandemic (2020 through 2022), but there has been a solid increase since 2022. The estimate in 2025 (641,000) is 26% higher than the estimate in 2013 (510,000).
31 CMAs, as a group
29.8 million people, or 72% of Canada’s population, reside in the 31 CMAs with reliable data on workers in the arts, culture, heritage, and sport.
84% of all Canadian workers in the arts, culture, heritage, and sport (roughly 550,000 people) reside in the 31 CMAs with reliable data.
Between 2013 and 2025:
The 28% increase in the number of workers in the arts, culture, heritage, and sport in the 31 CMAs is slightly above the Canadian average (26%).
29 of the 31 CMAs saw an increase, and the other two saw a very small decrease.
6 largest CMAs
20 million people, or 48% of Canada’s population, reside in the 6 largest CMAs.
62% of all Canadian workers in the arts, culture, heritage, and sport (roughly 408,000 people) reside in the 6 largest CMAs.
Between 2013 and 2025, the 26% increase in number of workers in the arts, culture, heritage, and sport in the 6 largest CMAs is equal to the Canadian average. Collectively, in the 6 largest CMAs, there were 314,000 workers in this occupation grouping in 2013 and 395,000 in 2025.
The following graph provides full details.
All 6 large CMAs saw increases in the number of workers in the arts, culture, heritage, and sport. Three had increases above 50%, while the other 3 had increases of 15% or less. Details follow for each CMA, from largest to smallest.
Toronto CMA
In the Toronto Census Metropolitan Area (population of 7.1 million in 2025), the number of workers in the arts, culture, heritage, and sport saw inconsistent increases between 2013 and 2025. Before 2025 (136,000 workers), the highest estimate was in 2017 (129,000 workers). The estimate was 118,000 in 2013, for an increase of 15% between 2013 and 2025.
Montreal CMA
Between 2013 and 2025, the number of workers in the arts, culture, heritage, and sport in the Montreal CMA (population of 4.6 million in 2025) increased fairly consistently, in relatively small increments. There were 88,000 workers in this occupation grouping in 2013 and 101,000 in 2025, representing a 15% increase.
Vancouver CMA
In the Vancouver CMA (population of 3.1 million in 2025), the number of workers in the arts, culture, heritage, and sport increased significantly between 2013 and 2025, with particularly large increases since 2022. There were 47,000 workers in this occupation grouping in 2013 and 78,000 in 2025, representing a 65% increase (the largest increase among the 6 large CMAs).
Calgary CMA
The number of workers in the arts, culture, heritage, and sport in the Calgary CMA (population of 1.8 million in 2025) increased by 60% between 2013 and 2025, with much of the increase coming in 2024 and 2025. The estimates of workers in this occupation grouping were 19,000 in 2013 and 30,000 in 2025.
Ottawa-Gatineau CMA
In Ottawa-Gatineau (population of 1.7 million in 2025), the number of workers in the arts, culture, heritage, and sport showed relatively little change between 2013 and 2023, with the exception of a pandemic downturn from 2020 to 2022. The estimate in 2025 (28,000) is 3% higher than the estimate in 2013 (27,000). The 3% increase is the smallest increase among the 6 large CMAs.
Edmonton CMA
In the Edmonton CMA (population of 1.7 million in 2025), the number of workers in the arts, culture, heritage, and sport increased quite consistently between 2013 and 2025. There were 14,000 workers in this occupation grouping in 2013 and 22,000 in 2025, representing a 53% increase.
Data source and limitations
Statistics Canada. Table 14-10-0468-01. Employment characteristics by census metropolitan area, annual. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1410046801
The Labour Force Survey (LFS) provides timely and accurate estimates, particularly for employed and unemployed workers in the overall labour force, where monthly data are available and reliable.
Given the relatively small sample size of the monthly survey (56,000 households), data on smaller population groups have higher margins of error. To counter this limitation, my analysis focuses on three-year averages (not annual or monthly data).
Other important limitations of the LFS:
People’s description of their main job during the reference period is used to classify them into occupations. Some data on secondary occupations is available but is not covered in this report.
It doesn’t have a large enough sample size to delve into the details for individual occupation groupings in Census Metropolitan Areas, which is why the broad category is helpful.
It captures data on Indigenous workers only every three months and only started capturing data on racialized workers in the summer of 2020.
No data are available for 2SLGBTQIA+, D/deaf, and disabled workers.
The Canadian census is a more robust source of information, including intersectional information and data on self-employment incomes. However, the census only takes place every five years.
Neither the LFS nor the census can provide insights into the reasons behind statistical trends.
A more thorough discussion of LFS limitations is available here.