Products, industries, or both? A re-analysis of economic impact statistics
Thoughts on provincial data and a reminder of the importance of methods and context
Today I am revisiting the topic of broad economic statistics on the arts and culture, a topic that I explored in a series of posts on macroeconomic indicators for the arts and culture, New information has come to light! Not new statistics, but some interesting things I’ve learned since those national posts.
Today’s article is a deep dive into the methods and findings of economic impact datasets. In particular, I reexamine provincial economic impact statistics.
My main point is that context is really, really important. A lack of context, in my view, is a big reason why some people believe that there are “lies, damn lies, and statistics”.
My analysis covers all provinces, but some parts of the article delve into the curious case of New Brunswick, a province in which I’ve been lucky enough to do a fair amount of work recently. In December, we released a final report on the impacts and challenges of cultural venues, which offers a summary of findings regarding venues’ financial health, human resources, community engagement, perspectives on Truth and Reconciliation, support for artists, attendance and marketing activities, physical condition, and perspectives on equity, diversity, and inclusion.
My thanks go out to Arts Link NB, the Owens Art Gallery, and the Association acadienne des artistes professionel.le.s du Nouveau-Brunswick for commissioning that work and for reviewing a draft of this post.
I should reiterate that data for other provinces are also included in this article and that the implications of my findings apply across the country. They could also apply to all types of measurements, not just ones related to economic impact.
Counting cultural products and/or cultural industries
My recent analyses of the economic impacts of culture, both Canada-wide (September-November) and in the provinces (back in July), focused on the “product perspective” available from Statistics Canada. Here is how I’ve explained that method and my decision:
The estimates relate to cultural products, i.e., the production of goods and services from cultural organizations and industries as well as non-cultural ones. The product-based estimates, which are typically a bit smaller than estimates of cultural industries [Note: I will revisit the phrase “a bit” in today’s article], are the focus of this article, simply because they have become more commonly used in the sector.
The other option would be to rely on the industry perspective, which has clear links to the product perspective: cultural industries are defined as ones in which the production of culture is a primary aspect of operations. But the production of culture is not the only thing that organizations, businesses, and individuals in cultural industries do.
Cultural organizations, businesses, and individuals also host fundraisers, rent their spaces for events, rent parking spaces, sell beverages and snacks, and much more. These other activities – and maybe you can think of some better examples – are counted in the industry perspective but not in the product perspective. (These will largely be referred to as “non-cultural products” below.)
On the other hand, the industry perspective excludes cultural production from organizations, businesses, and individuals that are not predominantly focused on cultural production. The crafts, performances, films, books, magazines, etc. that are produced in non-cultural industries are excluded from the industry perspective but included in the product perspective.
I did a quick but deep data dive on this topic and found that, according to Statistics Canada, nearly 90% of the crafts produced in Canada come from outside of industries that are predominantly focused on cultural production. This dwarfs the next-largest percentage: 22% of performing arts are produced outside of the cultural industries.
Sadly, or happily, there is no one “right answer”
Statistics Canada provides estimates of jobs, output (total revenues), and direct impact on GDP for both perspectives. When asked “what is the economic impact of culture”, you could reasonably lean on either method of calculation. Neither is more “right” than the other.
The choice depends on context. What do you want to achieve? If you want to compare the size of culture with other industries, then the industry perspective is probably better. If you want to talk about the importance of producing culture, then the product perspective might be the better choice.
Until this year, I had tended to analyze statistics using the industry perspective more often than the product perspective, because of the comparative aspect to other industries. Statistics Canada noted that the industry-based statistics are more comparable to how other industries are defined, not long after this impressive dataset was developed (emphasis added):
“The product perspective is unique to the [Provincial and Territorial Culture Indicators] and [Provincial and Territorial Culture Satellite Account], and therefore, there is no comparable figure for the economy as a whole. The industry perspective of the [Provincial and Territorial Culture Indicators] is more comparable to GDP by industry.”
However, for some reason, the product perspective has become more commonly used in the sector. Maybe this is because culture isn’t always like other industries, and we want (or need) a measurement of cultural production on its own, separate from the industries that are predominantly focused on its production, because of the uniqueness of the production of a work of visual art, a book, a musical performance, or a film.
Digging into the data
The two perspectives are strongly related but still very different.
Worded differently than the above description, the industry perspective includes:
Cultural products produced by cultural organizations, businesses, and individuals. This accounts for the lion’s share of the total revenues of cultural organizations, businesses, and individuals in Canada: 72% in 2022.
Non-cultural products produced by cultural organizations, businesses, and individuals (the other 28% of total revenues).
The product perspective includes:
Cultural products produced by cultural organizations, businesses, and individuals. This also represents the lion’s share of the total revenues related to cultural production in Canada: 83% in 2022.
Cultural products produced by non-cultural organizations, businesses, and individuals (the other 17% of total revenues).
If you are super-curious, here are lists of the products and industries considered cultural. Beware: many of the categories are quite vaguely worded, and the cultural elements are only a portion of the categories.
Variations across the country, and the curious case of New Brunswick
The industry perspective results in larger overall estimates of revenues (15% larger nationwide), direct impact on GDP (24% larger), and jobs (7% larger). These differences have increased over time and are now probably more than “a bit” larger, so I will retire that phrase from my description.
The differences between the two sets of estimates vary significantly across the country. The following graph shows per capita revenues based on the two estimates. The percentage differences between the industry perspective and the product perspective can be calculated by dividing the larger (industry) estimate by the smaller (product) estimate, and then subtracting 1 to get the percentage difference. (You also need to shift the decimal point by two spots, so that a calculation of 0.81 would become 81%.)