Government investment in culture: Comparisons with other indicators
Exploring comparisons with GDP, total revenues, other sectors, and changes over time
Today’s post provides a much requested but exploratory analysis of how government investment in the arts, culture, and heritage compares with Gross Domestic Product and output in the cultural sector. In addition, I explore how culture compares with a few other economic sectors, as well as whether these "return on investment" indicators increased or decreased since 2010.
The article focuses on nationwide data, mainly because federal investments in culture are not allocated by province in the dataset. This means that provincial calculations would be incomplete and very different from the national ones.
The math in today’s post is quite simple, but the context is tricky. I am calling this an “exploratory” effort, because I’m not 100% sure that the math is complete and appropriate. This series of reports is my first time working with the government spending dataset, and I am not yet sure that the government spending data are based on exactly the same definition of culture as the GDP and output data. As a result, I have a bit of uncertainty as to whether the “return on investment” ratios are based on a complete accounting of government spending.
One specific case in point: there is $2.8 billion in government spending on “Recreation, culture, and religion, not elsewhere classified” in 2023, which has not been included in my calculations of government spending on culture, because I have no way of knowing the cultural proportion of this spending. If at least some proportion of this amount were included, the return on investment ratios would be lower. I decided to offer “minimum” and “maximum” ratios in the cultural sector, based on including or excluding all of this spending,
Another tricky aspect involves the comparisons with other economic sectors. In addition to choosing sectors with definitions in the government spending data that appear to match the GDP data, I chose economic indicators that I believe to be the most appropriate comparisons for the custom-built economic indicators on the cultural sector.
Here is what I have done to alleviate my concerns:
I re-read the definitions included in the government spending dataset. The government spending data appear to be based on a thorough accounting of spending on culture. However, there is no mention of areas such as architecture or design in the government spending definitions. I am slightly reassured by the fact that I didn’t find architecture or design anywhere else in the spending definitions (outside of the cultural elements).
If there are aspects of government spending on culture that are excluded from the dataset but that should be included, the current ratios of return on investment would be too high.
The comparison sectors in this post were chosen because their definitions also appear to be the same between the government spending and GDP datasets. Once again, however, I am not fully confident in this. The comparison sectors are also ones that are not primarily “social” in nature, such as health, education, and social services. These areas account for significant amounts of government spending.
Because Statistics Canada offers estimates of the direct impact of culture on GDP expressed in terms of basic prices, I chose the other GDP indicators measured using basic prices, rather than market prices. GDP at basic prices is lower than GDP at market prices, mainly because basic prices exclude taxes.
Another note of caution: I sometimes hesitate to offer the types of simple ratio calculations that are provided in this article. Why? Mainly because there is no guarantee that an additional $1 in government investment will generate the same returns as in the past, in all locations and contexts. I do examine the historical range of these ratios in this article, to see whether and how they have changed over time.
I should note that the government spending data only include direct financial support, not tax credits (e.g., for) or other indirect instruments. This favours the cultural sector, because credits for film production and charitable donations (for example) are excluded from the spending totals, which results in higher ratios. However, I don’t know whether and how this exclusion affects the comparison sectors chosen for this article.
The data source and other important notes are included at the end of this post.