Skip to content

Canada · Publishing costs · 2022–2024

Higher employee compensation, lower revenue share: Canadian publishing costs reconciled

Employee compensation rose CAD 9.4 million, yet its revenue share fell 0.133 points. Other operating expenses rose CAD 92.3 million and their revenue burden increased 3.354 points. A reproducible accounting bridge separates dollar growth, denominator changes and the official rounded margin.

Published
Share
Share this page
EmailCite & reuse this study
Key findings as text
Reusable imagesHigher employee compensation, lower revenue share: Canadian publishing costs reconciled — +CAD 9.4m: Increase in employee compensation, 2022–2024. Nominal aggregate compensation and benefits, not a wage-rate or author-income estimate.Download sharing image (JPG) ↓

Key findings

+CAD 9.4m

Increase in employee compensation, 2022–2024

Nominal aggregate compensation and benefits, not a wage-rate or author-income estimate.

−0.133 pp

Employee compensation’s change relative to operating revenue

Recomputed from released rounded national amounts; an accounting ratio, not a causal effect.

+3.354 pp

Increase in other operating expenses per revenue dollar

Residual operating expenses outside the employee measure; not exclusively non-labour or production costs.

What the comparison shows

Canada’s book publishers reported higher operating revenue in 2024 than in 2022, yet the difference between revenue and expenses became smaller. Statistics Canada already reported this result in its February 2026 release. The useful next question is how to read the change without confusing a growing cost in dollars with a growing cost relative to the business’s revenue.

Using the current released national estimates, employee compensation increased by CAD 9.4 million between 2022 and 2024. Its ratio to operating revenue nevertheless decreased by about 0.133 percentage points. Other operating expenses increased by CAD 92.3 million, and their ratio to revenue rose by about 3.354 points. The two changes combine to a 3.221-point reduction in the margin recomputed from the rounded public revenue and expense figures. These are accounting relationships between industry estimates. They do not establish which decisions caused the change or what an individual publisher should spend.

The distinction is useful to writers, editors and publishing reporters because cost discussions often change denominator halfway through a sentence. A payroll can increase while taking a smaller share of revenue. A category can be the largest expense without contributing the most to the latest change. A margin can decline while sales or revenue grow. This reference provides a consistent way to keep all three observations visible.

What is being compared

The analysis uses the Canada rows of Statistics Canada Table 21-10-0200-01, Book publishers, summary statistics. The selected reference years are 2016, 2018, 2020, 2022 and 2024, all taken from the release available on 31 October 2026. The primary comparison is 2022 to 2024. The earlier observations provide descriptive context; they are not a continuous annual series or a causal baseline for an event.

Amounts are nominal Canadian dollars expressed in millions. They have not been adjusted for inflation. The target industry is book publishing, rather than all people who write books or every transaction in the retail book market. A publisher’s operating revenue is not identical to the money consumers spend at bookshops, an author’s royalty statement, or the net sales of publishers’ own and agency titles in the separate detailed tables.

The most recent estimates are preliminary and may change. The official release notes that 2020 and 2022 were revised. Keeping every observation from the same current table avoids assembling a historical series from releases that may contain different vintages. It does not make the data immune to future revision. A later refresh should rerun the whole selected series and retain the earlier snapshot for comparison.

A three-part account of every revenue dollar

Write operating revenue as R, operating expenses as E, and salaries, wages, commissions and benefits as W. The residual E minus W is labelled other operating expenses. This is an analytical grouping, not a new category that Statistics Canada collected independently. It includes all operating expenses outside the specified employee-compensation measure. It must not be called production cost, non-labour cost or discretionary spending: subcontracting and other payments for work can sit outside the employee measure.

Divide each component by R. For every CAD 100 of operating revenue, W divided by R gives employee compensation, E minus W divided by R gives other operating expenses, and R minus E divided by R gives the operating surplus. Multiplying each ratio by 100 expresses the values per CAD 100 of revenue. By construction, the three components sum to 100, apart from any final display rounding.

This shared denominator makes comparisons easier to audit. It also changes the question. A chart of expense composition asks how operating expenses are divided among categories. A chart per CAD 100 of revenue asks how much of the revenue base each category absorbs. The source’s separate industry-expenditure table uses a percentage-of-expenses basis. Its figures should not be placed on a revenue-burden chart without the required conversion and a clear explanation of the different denominator.

The dollar bridge from 2022 to 2024

The public figures show operating revenue increasing from CAD 1,649.1 million to CAD 1,700.7 million, a gain of CAD 51.6 million. Operating expenses increased from CAD 1,499.5 million to CAD 1,601.2 million, a gain of CAD 101.7 million. Subtracting expenses from revenue gives CAD 149.6 million in 2022 and CAD 99.5 million in 2024. The derived difference therefore decreased by CAD 50.1 million.

There is no forecasting model behind that last figure. The change in the surplus equals the change in revenue minus the change in expenses: 51.6 minus 101.7 equals minus 50.1. A bridge chart can start at the earlier surplus, add the revenue increase, subtract the expense increase, and finish at the later surplus. This makes it possible to check a visual against the underlying arithmetic rather than treating the chart as a separate source of evidence.

Splitting the expense change adds another layer. Employee compensation rose from CAD 372.8 million to CAD 382.2 million, or CAD 9.4 million. Other operating expenses rose from CAD 1,126.7 million to CAD 1,219.0 million, or CAD 92.3 million. The two changes sum to the CAD 101.7 million increase in total expenses. This decomposition describes the released aggregates, not the costs of the same fixed panel of publishing businesses.

Why a larger payroll can carry a smaller burden

Employee compensation grew more slowly than operating revenue in this two-year comparison. Its revenue ratio consequently fell. Recomputing the ratios at full precision from the released cells gives approximately CAD 22.606 in compensation per CAD 100 of revenue in 2022 and CAD 22.473 in 2024. The difference is about minus CAD 0.133 per CAD 100 of revenue, or minus 0.133 percentage points.

Other operating expenses moved in the opposite direction. Their ratio increased enough to absorb both the small reduction in the employee-compensation ratio and the decline in the derived margin. In symbols, the change in the margin is the negative of the change in the compensation burden minus the change in the other-expense burden. Here that is minus negative 0.133, minus 3.354, which yields approximately minus 3.221 points.

This calculation does not mean employees became cheaper, that workers accepted lower pay, or that compensation pressure disappeared. An industry total can change because of employment, hours, wage rates, benefits, business composition, entry and exit, classification, reporting or imputation. The aggregate cells do not separate those mechanisms. In particular, a compensation-to-revenue ratio is not a wage index and cannot establish what happened to the pay of a typical editor or production worker.

What the official margin and the recomputed margin each show

Statistics Canada publishes the operating profit margin rounded to one decimal place. It is 9.1% for 2022 and 5.8% for 2024. Those displayed numbers differ by 3.3 percentage points. Recomputing the ratio from the published revenue and expense values yields a decrease of about 3.221 points. Both descriptions can be appropriate if their origin is explicit. They should not be silently substituted for one another.

The apparent discrepancy follows from rounding and the fact that public aggregate amounts do not retain the source’s full calculation precision. The downloaded data keep the official margin, the recomputed margin and the difference between them in separate columns. The chart of the accounting decomposition uses the recomputed ratio consistently, because it must reconcile with its own public input amounts. A quotation of the official headline should retain the source’s displayed 9.1% and 5.8% instead.

Precision is not accuracy. Reporting six decimal places makes an identity easy to check, but it does not imply that an industry margin is measured to six decimal places. Public text should normally use one or two decimal places for the derived ratios and retain additional digits in downloadable calculations for reproducibility. The source estimates remain subject to sampling and non-sampling errors that this arithmetic does not remove.

A limited check on serialization uncertainty

Each selected monetary cell is displayed to one decimal place in millions. Under ordinary nearest-decimal rounding, the underlying rounded-to figure is within approximately CAD 0.05 million of the displayed amount. For a difference between revenue and expenses, allowing both cells to vary independently gives an outer rounding allowance of CAD 0.1 million. For a change between two such differences, the corresponding outer allowance is CAD 0.2 million.

This is a check on the consequences of published decimal precision, not a confidence interval. The calculation does not have the covariance information needed for a design-based interval for the margin change or the component burdens. It also does not claim that the public cells were rounded by a particular internal computer rule at every half-decimal tie. The outer allowance is intentionally conservative for reading the serialized monetary cells.

The CAD 50.1 million decline in the recomputed surplus is much larger than that rounding allowance. That observation establishes only that ordinary decimal serialization cannot explain the arithmetic sign of the change. It does not establish statistical significance, a causal effect, or a prediction about the next release. Sampling, administrative-data integration, business coverage and revision remain separate sources of uncertainty.

Employee compensation is a specific measure

The source definition includes salaries, wages, commissions and benefits for employees for whom a T4 slip was completed, together with specified employer contributions. It excludes working owners’ dividends and the remuneration of owners of unincorporated businesses. This matters in a sector containing owner-operated firms and independent creative work. The employee-compensation total is not a complete measure of everyone’s labour income associated with publishing.

Similarly, the operating margin has specific exclusions. Operating revenue excludes investment income, capital gains and certain non-recurring items. Operating expenses exclude items including write-offs, capital losses, interest on borrowing and certain non-recurring losses. The source explains that the operating surplus of unincorporated businesses includes unpaid remuneration to partners and proprietors that is not recorded as employee compensation. Calling the residual pure investor profit would therefore overstate what the measure identifies.

These boundaries also explain why the chart cannot answer an author’s personal pricing question. It combines many kinds of publishing businesses and activities. It contains no book-level advances, royalty rates, print quantities, advertising conversions, platform fees, return rates or title-level rights costs. A publisher can have a different cost structure without being inefficient, and an individual book can succeed or fail while the industry total moves in either direction.

Using the earlier years without inventing a trend

The five selected observations show how the accounting components vary across the current-release series. They are useful for checking whether a statement about 2022 and 2024 also holds at other observed points. They should not be joined into an annual record that implies values were measured in the intervening years. A straight line between two points is a visual connector, not an estimate for the missing calendar years.

The source notes a NAICS code replacement beginning with reference year 2022. Earlier years are retained as released in the current table, with that notice visible. The primary two-year bridge remains within the newer code period. The analysis does not claim to have proved complete economic comparability across the classification change or across the other survey-method changes described in the metadata.

The survey’s reference period can be a calendar year or a qualifying twelve-month fiscal period. It is not necessarily an identical January-to-December window for every business. The estimates combine survey responses and administrative information, with weighting, editing and imputation. They describe the industry domain for each period rather than tracking a fixed set of publishers through time. Changes in business composition can therefore contribute to changes in the aggregate ratios.

How to reuse the results responsibly

A defensible description is that, in the released national estimates, employee compensation increased in dollars between 2022 and 2024 while declining slightly relative to operating revenue, and other operating expenses account arithmetically for the increase in the expense burden. A defensible chart names the reference years, nominal currency, source table, current-release snapshot and preliminary status of 2024. It identifies the residual grouping and distinguishes recomputed ratios from the official rounded margin.

The accompanying tables retain the source vectors and coordinates so that a reader can locate each input. Reproduction requires no account or commercial data subscription: the public matrix contains the selected cells, and the code applies the stated identities. An independent check should compare those cells with the official downloaded table before checking the formulas. Repeating arithmetic alone would not catch a mislabeled universe or the accidental use of a different revenue measure.

The broader lesson is practical. Before describing an expense as a growing burden, identify both the amount and the denominator. Before describing the residual as profit, read its accounting definition. Before comparing two published percentages, check whether they share a base and a data vintage. Those habits make publishing statistics more useful to the people who need to interpret them, without asking the figures to answer questions they were not designed to measure.

Adapted from Statistics Canada, Book publishers, summary statistics, 2016, 2018, 2020, 2022 and 2024. This does not constitute an endorsement by Statistics Canada of this product.

Where each CAD 100 of publisher revenue goes

Canada book publishers · current-release estimates · nominal values · 2024 preliminary Derived ratios use released rounded revenue and expenses; they do not replace the official margin.

Full chart and source
Canada book publishers · current-release estimates · nominal values · 2024 preliminary
Derived ratios use released rounded revenue and expenses; they do not replace the official margin.

Scroll horizontally to inspect the chart. ↔

Chart and values

Scroll horizontally to see all columns.

Released amounts and a common revenue denominator
YearRevenue (CAD m)Expenses (CAD m)Employee compensation (CAD m)Other operating expenses (CAD m)Derived operating surplus (CAD m)Official margin (%)Recomputed margin (%)Employee compensation / revenue (%)Other expenses / revenue (%)
20161540.51389.9365.41024.5150.69.89.77623.72066.504
20181565.61456.9356.61100.3108.76.96.94322.77770.280
20201519.31366.6335.91030.7152.710.010.05122.10967.840
20221649.11499.5372.81126.7149.69.19.07222.60668.322
20241700.71601.2382.21219.099.55.85.85122.47371.676

Canada; nominal amounts; 2024 preliminary. Other expenses is E−W. Derived ratios use released rounded amounts; official margins remain separate.

Canada book publishers · current-release estimates · nominal values · 2024 preliminary Derived ratios use released rounded revenue and expenses; they do not replace the official margin.

Download chart (SVG) ↓ · Download chart (PNG) ↓

A higher compensation total can take a smaller revenue share

Canada · 2022–2024 · accounting decomposition, not causal attribution Margin change is the negative sum of the two expense-burden changes. No design-based interval established.

Full chart and source
Canada · 2022–2024 · accounting decomposition, not causal attribution
Margin change is the negative sum of the two expense-burden changes. No design-based interval established.

Scroll horizontally to inspect the chart. ↔

Chart and values

Scroll horizontally to see all columns.

Changes between adjacent selected reference years
FromToΔ Revenue (CAD m)Δ Expenses (CAD m)Δ Employee compensation (CAD m)Δ Other operating expenses (CAD m)Δ Derived operating surplus (CAD m)Recomputed margin change (pp)Compensation burden change (pp)Other-expense burden change (pp)
2016201825.167.0−8.875.8−41.9−2.833−0.9423.775
20182020−46.3−90.3−20.7−69.644.03.108−0.668−2.439
20202022129.8132.936.996.0−3.1−0.9790.4970.482
2022202451.6101.79.492.3−50.1−3.221−0.1333.354

Accounting identities, not causal attribution. Changes are nominal. No design-based confidence interval established.

Canada · 2022–2024 · accounting decomposition, not causal attribution Margin change is the negative sum of the two expense-burden changes. No design-based interval established.

Download chart (SVG) ↓ · Download chart (PNG) ↓

Data & downloads

Scroll horizontally to see all columns.

Released amounts and a common revenue denominator
YearRevenue (CAD m)Expenses (CAD m)Employee compensation (CAD m)Other operating expenses (CAD m)Derived operating surplus (CAD m)Official margin (%)Recomputed margin (%)Employee compensation / revenue (%)Other expenses / revenue (%)
20161540.51389.9365.41024.5150.69.89.77623.72066.504
20181565.61456.9356.61100.3108.76.96.94322.77770.280
20201519.31366.6335.91030.7152.710.010.05122.10967.840
20221649.11499.5372.81126.7149.69.19.07222.60668.322
20241700.71601.2382.21219.099.55.85.85122.47371.676

Canada; nominal amounts; 2024 preliminary. Other expenses is E−W. Derived ratios use released rounded amounts; official margins remain separate.

Scroll horizontally to see all columns.

Changes between adjacent selected reference years
FromToΔ Revenue (CAD m)Δ Expenses (CAD m)Δ Employee compensation (CAD m)Δ Other operating expenses (CAD m)Δ Derived operating surplus (CAD m)Recomputed margin change (pp)Compensation burden change (pp)Other-expense burden change (pp)
2016201825.167.0−8.875.8−41.9−2.833−0.9423.775
20182020−46.3−90.3−20.7−69.644.03.108−0.668−2.439
20202022129.8132.936.996.0−3.1−0.9790.4970.482
2022202451.6101.79.492.3−50.1−3.221−0.1333.354

Accounting identities, not causal attribution. Changes are nominal. No design-based confidence interval established.

Evidence & method

How we calculated it

  1. Select the 20 Canada cells for revenue, expenses, employee compensation and the published margin in 2016, 2018, 2020, 2022 and 2024 from current Table 21-10-0200-01. Retain vectors, coordinates, units and blank released statuses.
  2. Compute other operating expenses as E minus W, derived surplus as R minus E, and each component per CAD 100 of operating revenue. Keep the source’s independently rounded margin separate.
  3. For adjacent selected observations reconcile ΔP=ΔR−ΔE and Δmargin=−Δcompensation burden−Δother-expense burden. The primary comparison is 2022–2024.
  4. Run the supplied standard-library recipe against its pinned selected-cell matrix. Results JSON, source matrix and two CSVs reproduce byte for byte with required source notices. Decimal-rounding allowances are not confidence intervals.

What the data cannot tell us

  • Nominal aggregate estimates are not wages per worker, author income, book-level profit, prices, a stable publisher panel or a causal explanation.
  • Employee compensation excludes working owners’ dividends and remuneration of unincorporated business owners. The source margin can include unpaid proprietor remuneration.
  • Other operating expenses is a residual grouping that can include subcontracted work. It must not be relabelled non-labour, production or paper costs.
  • 2024 is preliminary and earlier data were revised. The primary comparison stays within the post-2022 NAICS code period; earlier years are descriptive context.
  • No survey covariance sufficient for design-based intervals was established. Decimal serialization sensitivity does not address sampling, nonresponse, classification, imputation or revision.

Original sources

  1. Statistics Canada: Book publishers, summary statistics (21-10-0200-01)

    National operating revenue, expenses, employee compensation and the independently rounded official margin; 20 selected cells.

    Accessed October 31, 2026

  2. Statistics Canada: 2024 book-publisher questionnaire

    Actual 2024 questionnaire: own and agency titles, net sales and the nationality/co-author/translator classification rules.

    Accessed October 31, 2026

  3. Statistics Canada: Book-publisher survey methods

    Survey methods: industry coverage, weighting, administrative information, fiscal periods, imputation and uncertainty.

    Accessed October 31, 2026

  4. Statistics Canada: Book publishing industry, 2024

    The official release already reports headline growth, margins and authorship results; these are prior findings, not discoveries of this analysis.

    Accessed October 31, 2026

  5. Statistics Canada Open Licence

    Source reuse terms require attribution and nonendorsement notices; no agency insignia or confidential data are used.

    Accessed October 31, 2026

Cite & reuse this study

Alisson Bento. 2026. Higher employee compensation, lower revenue share: Canadian publishing costs reconciled. Original secondary analysis of Statistics Canada’s current released publishing tables. https://alissonbento.com/research/canada-book-publisher-margin-cost-bridge-2016-2024/

Adapted from Statistics Canada, Book publishers, summary statistics, 2016, 2018, 2020, 2022 and 2024. This does not constitute an endorsement by Statistics Canada of this product. Original analysis and charts: Alisson Bento. Source information remains under the Statistics Canada Open Licence; original reproduction code is MIT-licensed. No agency endorsement or licence for third-party book content is implied.

Corrections or editorial questions

Pass it on.

Higher employee compensation, lower revenue share: Canadian publishing costs reconciled | Alisson Bentoalissonbento.com

WhatsAppFacebookXTelegramRedditThreadsEmailSMS