Box-Office Economics: How a Film Makes (or Loses) Money from Theatres to Streaming

Box-office economics is the arithmetic that decides whether a film is a business or a charity. Here the Kurums Startup team walks through where each dollar of a ticket goes, why a film that “grosses $300m” can still lose money, how China’s Ne Zha 2 rewrote the record books without Hollywood, and how streaming changed the back end. This article is part of the Arts & Creative Economy hub; if you want the financing side first, read how movies get funded, and for the distribution shift see streaming vs theatrical.

How big is the global box office and who is growing?

Worldwide cinema ticket sales were about $33.6bn in 2025, Gower Street Analytics reported in January 2026, roughly 10% above 2024 but a fifth below 2019. Its April 2026 revision puts 2026 at $34.7bn: slow recovery, not a rebound.

The recovery is uneven. North America ended 2025 at roughly $8.8bn, up about 1.5% on a weak 2024 according to Comscore figures reported by Screen Daily, still about 24% below the $11.4bn of 2019; the summer season produced only $3.67bn, one of the weakest non-pandemic summers in decades. Sequels underperformed; family animation and originals (Sinners, F1, Weapons) over-delivered.

China moved the other way. Its 2025 gross hit 51.8bn yuan (about $7.4bn), up almost 22% year on year, with domestic films taking 79.7% of revenue, per China Film Administration data reported by the South China Morning Post. One title, the animated Ne Zha 2, accounted for nearly a third of the entire year, grossing more than 15.9bn yuan (about $2.2bn) on the mainland and 324 million admissions according to Maoyan. Even so, 2025 China fell short of 2023’s 54.9bn yuan, a reminder of how dependent the market is on one or two Spring Festival hits.

Elsewhere, Japan’s Demon Slayer: Infinity Castle ($793m worldwide, over 80% outside North America) and Disney’s Zootopia 2 (past $1.87bn, three-quarters international) showed that the “domestic” share of the global pie is now below 30%, and the mix of markets matters as much as the headline number.

Where does the money from a cinema ticket actually go?

A ticket is split between exhibitor and distributor on a sliding scale: distributors take 60–70% in the opening weeks, falling to 30–40% later, averaging roughly half the domestic gross on a wide release. The cinema keeps the remainder plus nearly all food-and-beverage revenue.

Cinemark’s FY2025 annual report describes its film rental fees as sliding-scale formulas tied to box office receipts, and across US exhibition film rental and advertising costs run at roughly 55% of admissions revenue. Concessions are the exhibitor’s profit engine: Cinemark reported record food-and-beverage spend of $5.96 per guest in the fourth quarter of 2025 on product costs of about a fifth of that, across 44.3 million admissions, a high-margin stream the studio never touches.

Internationally the arithmetic changes. In the UK, Germany or France distributors typically see 40–45% of the gross after VAT and local levies (France’s CNC ticket levy is about 10.7%). In China, imported revenue-share films return only about 25% of the gross to the foreign studio under the quota system run by China Film Group, so a $100m Chinese gross is worth about $25m to a Hollywood studio.

Average ticket prices also differ widely: roughly $12 for top-ten US releases in 2025 per The Numbers, inflated by premium formats (Axios reported IMAX heading for a record $1.2bn of revenue), against $1.50–2.00 in India and $3–4 in Türkiye. Admissions, not dollars, measure a market’s cultural reach.

What does it cost to release a film, and why is marketing so expensive?

Releasing a studio film costs about as much again as making it. Prints-and-advertising (P&A) budgets of $100m-plus are standard for tentpoles, and even a $20m independent needs $10–15m of marketing to open wide in the US, which is why most independents open on a few hundred screens.

P&A covers paid media, trailer placement, publicity, premieres, dubbing and subtitling, and digital distribution fees. Studios spend heavily in the final three weeks because opening weekend typically generates 25–40% of total domestic gross; a film that opens softly rarely recovers. Warner Bros. reportedly spent around $100m worldwide marketing Superman (2025), which grossed about $619m, and comparable sums supported A Minecraft Movie ($960m) and Jurassic World Rebirth ($872m).

Distribution fees sit on top: a studio charges the production roughly 30% of domestic rentals and 35–40% internationally, before recouping P&A with interest. This is the mechanism behind “Hollywood accounting”: a film can gross $400m, return $200m in rentals, and still show a paper loss after fees, P&A, interest and overhead, which matters because many talent deals are defined against net rather than gross.

This cost structure is why studios chase sequels and pre-sold brands: marketing a known title is cheaper per dollar of gross, and the back end is larger. It is also why 2025’s originals, Sinners (about $280m domestic on a $90m budget) and Apple’s F1 (about $634m worldwide), were treated as events rather than the norm.

How do you calculate whether a film made money?

Start from worldwide gross, deduct the exhibitors’ share to get rentals, subtract P&A and the distribution fee, then compare the remainder with the negative cost (budget plus financing and bond costs). Add ancillary revenue over five to seven years, then subtract participations and residuals.

A worked example: a studio film with a $150m negative cost and $120m of global P&A grosses $250m domestic and $300m international. Domestic rentals at 52% are $130m, international at 40% are $120m, total $250m. Deduct P&A ($120m) and a blended 32% distribution fee ($80m) and the studio has $50m against a $150m negative cost: a $100m theatrical shortfall. PVOD and home entertainment might add $60m net, a pay-one streaming licence $70m, free TV $40m and international TV $30m, leaving a modest profit of around $50m before participations. That is the typical shape of a middling tentpole: theatrical loss, ancillary rescue.

Now change one input. If the same film grosses $1bn worldwide, as Lilo & Stitch did in 2025 on a reported budget of around $100m, rentals rise to roughly $450m and the film is profitable in theatres alone. Theatrical economics are convex: hits are far more profitable than misses are unprofitable, which is why studios tolerate a 60–70% miss rate on their slates.

Independent films use the same logic with a different ordering of claims: pre-sold territories pay minimum guarantees on delivery, the sales agent takes commission, the bank recoups, equity recoups with a premium, and only then do producers share profits. The film financing guide walks through that waterfall.

Why did Ne Zha 2 change the box-office conversation?

Ne Zha 2 became the first film to gross over $1bn in a single market and the highest-grossing animated film ever, with about $2.2bn worldwide and 97% of it from mainland China according to Maoyan data cited by Yicai Global. A national market produced a global top-five film without Hollywood.

The economics were unusual. Produced by Chengdu Coco Cartoon with backing from Beijing Enlight Media on a reported budget around $80m, a fraction of what Pixar spends, the film kept a far higher share of each yuan than any foreign title could, and Enlight’s share price roughly doubled after the Lunar New Year opening. IMAX reported it as its highest-grossing animated release ever.

The implications are mixed. Zootopia 2 set a record for an imported film in 2025, yet US studios’ aggregate share of the Chinese market is now below 20%, down from roughly 40% a decade ago. Omdia’s analysts have argued that Chinese animation could become a soft-power export in the way Japanese anime already is, with Demon Slayer’s $793m worldwide run as the template.

The cautionary point is concentration: without Ne Zha 2, China’s 2025 box office would have fallen rather than risen 22%, and with more than 80,000 screens the exhibitors’ problem is utilisation outside the Spring Festival peak, not capacity.

How long is the theatrical window and why does it matter?

The average exclusive theatrical window for the top 100 films of 2025 was 37 days, three days longer than in 2024, Cinema United’s Michael O’Leary told CinemaCon in April 2026. Exhibitors want a 45-day floor, and in 2026 the largest studios have moved towards it.

The window is how long a cinema has a monopoly on a title before premium video-on-demand (PVOD) at $20–30 per rental or streaming. Universal, which pioneered 17-day windows in 2020, moved to a 31-day minimum for its 2026 slate and a 45-day minimum from 2027, keeping Focus Features on the shorter structure. Paramount Skydance’s David Ellison pledged 45 days for all Paramount releases and 60–90 for its biggest titles. Disney has run close to 100 days since Avatar: The Way of Water, and Sony has released 29 films with 45-day-plus windows since 2022, per TheWrap.

Exhibitors point out that when audiences expect a film at home within three weeks, later weekends collapse faster; Five Nights at Freddy’s (2023), released day-and-date on Peacock, fell 76% in its second weekend. Studios counter that a short window lets one marketing campaign serve both theatrical and PVOD, and that PVOD rentals, split roughly 80/20 in the studio’s favour, carry far higher margins than a ticket. The 2026 compromise, longer exclusivity for tentpoles and flexible windows for mid-budget films, reflects the fact that both sides are right about different films.

Cinemark’s FY2025 10-K states that the window remains a significant driver of box office performance. For how value moved between windows, see Streaming vs Theatrical.

How do the economics differ for studios, independents and streamers?

Studios play a portfolio game: 15–25 films a year, most of which will not recover their cost in theatres, carried by two or three hits plus library income. Independents finance film by film with pre-sales and incentives; streamers pay fixed fees that remove box-office risk but cap the upside.

The table compares revenue flows under each model. The streamer “cost-plus” model, in which Netflix, Amazon MGM or Apple pay the budget plus a 10–20% premium and own all rights, has no box-office line at all, which is why Apple’s F1 (about $634m worldwide via Warner Bros.) was studied so closely: a rare case of a streamer treating cinemas as revenue rather than marketing.

Türkiye illustrates the independent model at national scale. Admissions have recovered to roughly half the pre-pandemic level of about 60 million according to Box Office Türkiye tracking, domestic comedies still take most of the gross, and the real engine of the Turkish screen economy is television drama exported to over 150 countries. A Turkish feature typically finances itself through a broadcaster or platform pre-sale, Ministry of Culture and Tourism support and a distributor advance, with theatrical gross treated as a bonus.

What should investors and founders take from box-office economics?

Three rules: the ticket is shared before anyone profits, marketing is a second budget, and the window decides how much of a film’s value the cinema captures versus the studio. Model all three rather than the headline gross.

For investors, prefer slates, funds or structured positions (tax-credit loans, pre-sale discounting, completion-bonded gap) over single-picture equity, and insist on seeing the collection account agreement, because the waterfall, not the gross, determines who gets paid. For founders of adjacent businesses (ticketing, analytics, premium formats, localisation), growth is in the parts of the chain that scale with admissions rather than hits: Cinema United noted in October 2026 that North American exhibitors have reinvested $2.7bn in their cinemas over two years.

For policy people, the national-champion model of China, Korea and Türkiye shows that box office is a lever of industrial policy. The incentives that attract production are covered in film tax credits by country, and the wider sector numbers in how big is the global creative economy.

Frequently Asked Questions