China's National Bureau of Statistics published its quarterly reading on large culture companies on 30 October, and the two headline numbers pull in opposite directions. Operating revenue for the first three quarters of 2026 came to 108,210 hundred million yuan — roughly 10.82 trillion yuan — up 4.3% year on year on a comparable basis. Total profit was 8,317 hundred million yuan, down 7.3%. The operating-revenue profit margin slipped to 7.69%, 0.96 percentage points below a year earlier.
That combination — more sales, less profit — is the part of the release worth arguing about. The release itself does not explain it.
A large-firm panel, not the whole sector
The survey covers about 83,000 culture and related-industry enterprises "above scale". The threshold depends on what the company does: industrial and wholesale enterprises qualify at 20 million yuan of annual main-business revenue, retailers at 5 million, service enterprises at 10 million, with carve-outs at 20 million for transport, warehousing and post, information transmission and software, and water and environment utilities, and at 5 million for resident services, repair, and culture, sport and entertainment.
So this is a panel of big operators, identified each quarter as legal units and aggregated from full returns on the national statistical reporting platform. It is not the sector, and it is not a fixed roster: firms cross the threshold, shrink out of it, are newly built, go bankrupt or deregister. Growth rates are therefore computed on a restated base consistent with the current period, which means they will not match levels published a year ago. Every rate in the release is nominal — no price adjustment.
Where the 8.9% comes from
The 16 subcategories flagged as new-format cultural businesses took in 53,827 hundred million yuan, up 8.9%, or 4.6 percentage points faster than the whole panel. That is just under half of total revenue, which makes the group too large to treat as a curiosity.
The basket is worth reading closely, because it is not a clean "digital media" measure. It includes internet search, other internet information services, internet advertising, internet games, internet cultural entertainment platforms and value-added telecom cultural services; digital publishing, multimedia/game-animation/digital-publishing software and copyright and cultural software; animation and game digital content, other cultural digital content and other culture and arts activities; integrated broadcast control; and three hardware lines — entertainment smart drones, wearable smart cultural devices and other smart cultural consumption devices. Advertising sales, software licensing and gadget manufacturing sit in the same 8.9%, so the number cannot be read as a single demand story.
By industry type, the split is blunt. Cultural services were the engine at 65,697 hundred million yuan, up 7.3% — about 61% of the panel's revenue. Cultural manufacturing was 26,744, down 0.2%, and cultural wholesale and retail 15,769, down 0.1%. The physical and distribution end did not collapse; it edged backwards while services grew.
By segment: content creation and production remained the largest at 25,129 (+4.1%); creative design services 20,850 (+8.8%); news and information services 16,337 (+7.3%); cultural consumer terminal production 16,383 (flat); auxiliary production and intermediary services 11,673 (+4.7%); cultural dissemination channels 10,780 (−1.9%), the only sizeable decline; cultural equipment production 5,112 (+5.3%); cultural entertainment and leisure services 1,535 (+0.3%); and cultural investment operations 412 (+11.7%, the fastest rate off the smallest base by a wide margin). The six core segments together produced 75,042 (+5.1%); the three related segments 33,168 (+2.4%).
The profit line
Assets at end-September were 235,484 hundred million yuan, up 6.0%. Revenue per 100 yuan of assets was 62.0 yuan, 1.6 yuan less than a year earlier. In plain terms, the balance sheet grew faster than the sales drawn from it, and the margin thinned at the same time.
The release gives no cost breakdown, no segment-level profit and no company detail, so it cannot tell a reader whether the squeeze came from labour, content rights, marketing, hardware pricing or anything else. Anyone who supplies a cause is going beyond the document.
The map
The east — Beijing, Tianjin, Hebei, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong and Hainan — booked 88,123 hundred million yuan, up 5.0%, or about 81% of the total. The centre was 11,165 (+1.5%), the west 8,044 (+2.0%), and the northeast 877 (−5.5%). The northeast, covering Liaoning, Jilin and Heilongjiang, was the only contracting region and by far the smallest base; a percentage move there is cheap to produce, but the direction is consistent with a thinner roster of large culture firms in those three provinces.
What the numbers can support
Three conclusions survive the caveats. First, large culture firms sold more and kept less: revenue up 4.3%, profit down 7.3%, margin down 0.96 points. Second, growth is concentrated in services, creative design and the new-format subcategories, while dissemination channels, manufacturing and wholesale/retail stalled or slipped. Third, the asset base is expanding faster than revenue, which is a question for the next few quarters rather than a verdict.
What the release cannot support is any claim about small firms, employment, prices or household spending on culture — none of that is in scope. Nor can it say whether the margin fall is a one-quarter wobble or a trend; nine-month cumulative figures hide the quarterly path. The watch list for the full-year release is short: whether the dissemination-channel decline persists, whether the margin stops falling, and how much of the new-format growth is software and services rather than hardware.