A few months ago I attended a virtual convening of non-profit arts leaders from across America gathering to discuss the arts model and how it might be strengthened. I wondered why we were just talking among the non-profit arts sector, when it seems to me the commercial creative industry, though bigger, is facing many of the same issues. And that led me to wondering about the comparative size of the non-profit and for-profit culture industries.
If we’re going to think about how to fix a culture model that’s not working well, we should try to understand the scale of what we’re actually talking about. So over the past few months I’ve been compiling sources to attempt to find out, and it turns out that it’s a project that requires not just hunting reliable numbers, but making a series of philosophical choices as to what and how you count. Last week I wrote about comparing the economic size of both sectors.
But that’s only part of the story. Dollar value is certainly one way of measuring (and what people choose to spend their money on is a good indicator of interest). But we should also consider audience size. And ideally, we’d like to measure the impact or engagement people have with the culture they choose. I’m not sure how to defensibly measure that last part, but we can take a stab at measuring audience, though it turns out doing that requires even more calibration in what and what not to include.
So I propose the following not as the definitive accounting, but my best attempt to understand the relative scale of the culture industries. I think the numbers suggest a different way of thinking about making a case for the traditionally non-profit arts sector.
By revenue, the non-profit sector is small — about $73 billion in organizational spending compared to $1.17 trillion in total US arts and cultural production. Disney’s annual revenue alone is larger than every US nonprofit cultural institution in the country combined.
But the map of audience shows something entirely different.
Americans visit museums roughly 850 million times a year, according to the AAM. They buy 800 million movie tickets, according to box office numbers. The aggregate scale of nonprofit and commercial cultural attendance turns out to be within shouting distance of one another — same order of magnitude, similar scale of participation.
Same audience, but vastly different economics.
The headline numbers, side by side: Total US non-profit live cultural attendance by my accounting (sources and methodology at the end) comes in around 965 million a year — museums (~850M), nonprofit theatre (27M), performing arts centers and presenting venues (~25M), orchestras (22M), nonprofit festivals like Tanglewood and Aspen (~20M), choral concerts (~10M), university presenters like Cal Performances and UMS (~8M), opera (2M). Total US commercial live cultural attendance comes in around 1.37 billion — movies (800M), theme parks (158M), the Big Four sports leagues combined (134M), college and minor-league sports (125M), Live Nation US concerts (60M), Broadway and national touring (29M), Las Vegas residencies and shows (~25M), comedy clubs and dinner theatre (20M), commercial music festivals (12M), the WNBA and NWSL combined (4.4M).
Roughly the same order of magnitude. The non-profit number comes in at about 70 percent of the commercial. Compare that to the revenue numbers — for-profit industry revenue is 16 times larger than the non-profits.
The composition is wildly different. Museums dominate the nonprofit number. Movies dominate the commercial. Live performing arts and presenting venues are small in either column but significant in both. By participation, the non-profit cultural sector is a near-peer of the entire commercial culture economy. So the dollar gap between the two isn’t a popularity gap, it’s a pricing gap.
This comparison is built on live, in-person, ticketed or admission-tracked cultural attendance. It excludes digital cultural engagement — streaming video, broadcast television, video games, podcasts, music streaming, social media. Americans average roughly five hours a day on video streaming alone, and adding screen time would push every commercial figure into the trillions of hours and reshape the comparison around something other than physical participation. That’s a different post for another time. Where category aggregates are published by a trade body or league, I’m using the trade body’s number; where none exists, I’m using documented bottom-up estimates with ranges. Methodology, source links, and cautions for the numbers are at the end of this post.
So what can we learn from this? The notion that non-profit culture is much smaller than for-profit because fewer people participate is flat wrong. In popularity, it turns out that pop culture industries aren’t that much more popular. In fact, museums are much more popular than many kinds of “pop” culture. When you look across the vast array of different cultures that Americans consume, it looks like a mosaic, admittedly anchored by museums and movies, but otherwise distributed across a wide range of interests. “Pop” culture, as we thought of it in the 20th Century in the TV era, as a monolith of popular choice, no longer dominates the national psyche. We have fragmented our attention, for good and bad, and we have many kinds of fame now, not just the pop culture celebrity version.
What sticks out in the comparison is the disparity in culture monetization. Per-attendee economics, in round numbers, are instructive. Disney’s domestic parks pull about $190 from each guest in admissions and in-park spending. An NFL game with concessions runs $150 or more. A movie ticket: roughly $11. A nonprofit theatre seat: $25 to $50. A museum visit: $0 to $25, average closer to $10 once you weight free admission and discounted hours.
So a single Disney park visit is monetized roughly twenty times as aggressively as a museum visit. An NFL game extracts about fifteen times what a museum visit does. Disney’s global parks-and-experiences division produced $34 billion in revenue last year off 140 million park visits and the surrounding Disney machine of hotels, cruises, and merchandise the parks anchor. Disney’s most-popular park, in Florida, attracted $17.8 million visitors. The Smithsonian drew 17 million visits and generated effectively zero visitor revenue. Both organizations are doing precisely what they were designed to do.
The commercial cultural economy isn’t bigger because more people show up. It’s bigger because each attendance gets monetized at a very different scale. This isn’t a moral judgment: extraction at scale is what for-profit models are built to do and they do it well. But it means the non-profit sector’s economic “smallness” is structural on purpose, not evidence of marginal cultural standing. The free museum, the priced-for-access regional theatre, the subsidized concert ticket — these aren’t bugs in the design, they are the design.
But here’s where I think the story gets dark. The frame for non-profits is rigged by the language. We call commercial culture “free-market capitalism” and non-profit culture “subsidized.” Both descriptions are wrong. Public dollars flow into the commercial cultural economy at massive volume. NFL and MLB stadiums have absorbed roughly $33 billion in taxpayer subsidy over the past three decades. Film and TV production tax credits cost US states about $25 billion a year. Stan Kroenke’s new $5 billion Rams stadium in Los Angeles got a reported $400 million in public subsidy. Georgia’s film tax credit alone puts $1.3 billion a year on the public ledger (shout out to “Walking Dead” fans).
These dollars don’t pay for the games or the movies. They reduce the capital and operating costs of for-profit enterprises whose profits flow to private owners — Kroenke, Disney shareholders, studio shareholders. The subsidy capitalizes the owners, with the public return the privilege of being allowed to buy a ticket on the owner’s terms.
Public dollars flow into the non-profit cultural sector in a fundamentally different way. NEA grants expand accessibility. State arts council appropriations subsidize ticket prices, fund commissions, support touring, and build infrastructure that’s permanently held in public trust. The Met Museum cannot be sold to a hedge fund (at least officially). The Smithsonian cannot be taken private. The mission is the property.
The same word, “subsidy,” is doing two different jobs. One capitalizes private wealth and routes the cultural output through paywalls. The other produces public goods that are legally bound to public purposes. Treating them as morally equivalent — or worse, pretending one is free-market, earning its way, while the other is government largess or charity — is just wrong. And yet that is the frame that the non-profit arts industry so often accepts in its funding arguments.
Here’s a thought experiment: what if the subsidies were reversed and government subsidized benefits to the cultural products that generated the most returns to the most people instead of a tiny number of billionaire NFL owners and oil company execs and Disney? A non-profit culture subsidy comparable to that for stadiums annually would quintuple the size of the NEA budget.
To be fair, we have to acknowledge that non-profits are subsidized by not paying taxes that commercial entities do. But resetting the language makes it clearer how the business frames work. There are examples of this subsidy distortion across the American economy. Oil companies earn billions of dollars a year in profits. But the industry is also heavily subsidized. As is agriculture.
So what?
- For arts policy and advocacy, the relevant unit of measurement isn’t dollars, it’s participation. Federal NEA appropriations plus all state arts agencies total about $965 million a year in service of a sector that delivers nearly a billion live cultural attendances. About a dollar of public investment per cultural participation. Compare that to the per-attendance public subsidy on a $400 million stadium hosting ten home games for 70,000 fans each. The math isn’t close.
- For non-profit cultural leaders, the move should be to stop fighting on the metric the model was specifically designed not to optimize for. The “we contribute $X to the economy” argument was never one the structure was built to win. The argument that’s actually defensible and accurate is that the non-profit sector delivers hundreds of millions of cultural participations a year at low cost per person served, in a form that cannot be replicated commercially without changing what it is. Argue infrastructure, not industry.
- The for-profit cultural sector is not the adversary. It faces the same structural pressures bearing down on the non-profit world, just at vastly larger scale. Audience attention is fragmenting. AI is flooding the content market. Platform intermediaries are extracting margin. Younger audiences are reconsidering what they’ll pay for live experience. The Met Opera and Live Nation are wrestling with the same demographic puzzle. Disney is fighting the same fragmentation problem regional theatres are. The shared question is what culture is for, who pays for it, and how the people who actually make it get paid.
- The adoption of AI will change
mucheverything about how the culture industry — non-profit and commercial — will work. It will scramble how value is assigned, who the audience is, how artists get paid, who has jobs, and how culture is delivered. Big Tech is throwing hundreds of billions of dollars into this makeover project with the expectation of extracting trillions of dollars. The thing is, those AI models are built on the knowledge and creativity and industry we humans have produced. We shouldn’t give that away for free; it would be the mother-of-all-subsidies if we did. But maybe if we reframe the way we understand how the culture economy works we will have a stronger claim in benefitting from the value extracted.
Free markets are policy. Cultural policy is policy. We’ve already decided that some forms of cultural infrastructure are worth public investment, at substantial scale. The question is whether the choices we’ve made reflect what we actually want our cultural infrastructure to be. The answer at the moment is probably not. By participation, the non-profit cultural sector is one of the largest broad-access cultural enterprises in the country. But the funding model treats it as a niche concern. That we should definitely change.
Methodology and sources
Building a credible head-to-head comparison of non-profit and commercial cultural participation in the US is difficult and messy. The two halves of the cultural economy report through different bodies, on different fiscal calendars, using different definitions. Some categories have rigorous trade-body aggregates while others have no national reporter at all. So herewith, some caveats about the count and some rationales.
What the numbers count
Live, in-person, ticketed or admission-tracked cultural attendance in the United States. This includes free admission where it’s tracked (most museums, most outdoor public concerts) and excludes physical activity that isn’t really cultural participation (mass transit, retail, religious services, National Park visits…).
What the numbers don’t count
Digital cultural engagement — streaming video, broadcast television, video games, podcasts, music streaming, social media. Adding any of these would push the commercial side into the trillions of engagement-hours and recast the comparison as a contest about screen time rather than physical participation. The case for live cultural infrastructure is fundamentally a case about civic gathering, and that’s the comparison this post measures. The digital-engagement question deserves its own treatment.
How I got the totals
Where a trade body or league publishes a national aggregate, I’m using the trade body’s most recent figure. Where no aggregate exists, I’m using midpoint estimates from bottom-up counts of documented major venues plus reasonable ranges for the long tail. Below I explain estimates.
By category
Firm — published trade body or league figures:
- Museum visits — 850M (American Alliance of Museums; figure predates pandemic and current is ~700–800M; long-cited as the canonical aggregate)
- Movie theater tickets — 800M (Motion Picture Association)
- Theme park visits — 158M (TEA Theme Index 2024, top 24 US parks)
- Big Four sports combined — 134M (MLB 71M, NFL 18M, NBA 22.5M, NHL 22.5M; league reports)
- College + minor league sports — 125M (NCAA football ~50M, NCAA basketball ~30M, MiLB 31M, MLS 12M, plus other minors)
- Live Nation US concerts — 60M (Live Nation 2024 10-K, Venue Nation US attendance only)
- Broadway + national touring — 29M (Broadway League, 2023–24 season)
- Nonprofit theatre — 27M (TCG Theatre Facts 2023)
- Orchestras — 22M (League of American Orchestras Impact Report 2023–24)
- Commercial music festivals — 12M (festival-level reporting summed)
- WNBA — 2.35M (record 2024 season; WNBA / ESPN)
- NWSL — 2.0M+ (record 2024 season; NWSL league reports)
- Opera — 2.1M (OPERA America 2022–23 Annual Field Report)
Estimated — midpoint range:
- Performing arts centers — ~25M (range 20–30M); bottom-up from documented major venues (Lincoln Center ~5M, Kennedy Center ~3M, Hollywood Bowl ~750K, Wolf Trap ~355K, plus 30–40 large regional PACs). No national aggregator publishes a sector total.
- Las Vegas entertainment — ~25M (LVCVA reports 41.7M total visitors with ~37% attending shows × ~1.5 shows each ≈ 23M; rounded to 25M for ticketed entertainment)
- Comedy clubs + dinner theatre — 20M (NEA SPPA shows 7–8% adult attendance × frequency)
- Nonprofit festivals — ~20M (range 15–25M); bottom-up from documented festivals (Tanglewood 350K, Ravinia 600K, Aspen 100K, Hollywood Bowl 750K, Grant Park ~227K, plus regional Shakespeare and music festivals)
- Choral concerts — ~10M (range 6–15M); bottom-up estimate from ~12,000 US choruses × 3 concerts/year × ~250 average audience. Chorus America reports 42M Americans participate in choruses as singers, but does not publish an aggregate audience figure. The 10M estimate captures concert audiences only, which is a different (smaller) population than the 42M who sing.
- University presenters — ~8M (range 5–10M); bottom-up from major university venues (UMS Michigan 150K, Cal Performances 250K, plus 100–150 mid-sized university presenters)
What I left out
These categories would push the nonprofit number higher if reliably aggregated: community theatre (the American Association of Community Theatre lists 7,000+ member theatres but does not publish an aggregate attendance figure; bottom-up estimates range from 7–15M), public library cultural programming, K-12 school arts attendance, National Parks Service historic and cultural site visitation (~40–60M annually), and free outdoor festival attendance not captured under the festival line. Including them is consistent with the framing of nonprofit cultural participation but unreliable.
These categories would push the commercial number higher: NCAA hockey and lacrosse, professional rodeo, dinner shows in non-Vegas markets, and a long tail of independent venue concerts. The Live Nation figure is one promoter’s US venue attendance; total US concert attendance across all promoters is meaningfully larger.
Caveats on specific figures
The museum 850M figure is AAM’s standard citation, originally from the early-2000s era of museum data collection. AAM’s 2024 surveys show roughly half of US museums have not yet returned to pre-pandemic attendance, so current attendance is probably closer to 700–800M. Using 850M is consistent with the figure museums and policymakers cite; using 750M is consistent with current AAM survey data. The story holds either way.
Performing arts numbers (theatre, orchestras, opera) reflect 2023 fiscal year data and are still recovering from the pandemic. 2024–25 figures will likely be modestly higher across all three categories.
The Disney $190 per-capita extraction figure is FY2024 domestic-park per-capita guest spending (admissions plus in-park food/beverage and merchandise) as reported in Disney’s 10-K. International parks per-capita is somewhat lower; globally weighted is approximately $170.
Sources
Nonprofit cultural attendance
Commercial cultural attendance
- Motion Picture Association — Theme Report
- TEA / AECOM Global Experience Index 2024
- League reports: MLB, NFL, NBA, NHL, MLS, MiLB, NCAA
- WNBA — 2024 record-setting season (2.35M total attendance)
- NWSL — 2024 attendance report (2.0M+ total attendance)
- Live Nation 2024 10-K (SEC filing)
- Broadway League — Season Statistics
- Las Vegas Convention and Visitors Authority — Visitor Profile
- Festival-level reporting (Coachella ~750K, Lollapalooza ~460K, Austin City Limits ~450K, EDC Las Vegas ~525K, Stagecoach ~75K, Outside Lands ~225K, Bonnaroo ~80K)
Per-attendee revenue figures (Section 3)
Subsidy figures (Section 4)
Reference data from Part 1
I realize there is a whole lot of judgment involved in creating this picture. My goal isn’t to definitively measure, but to try to build a case for the scale comparison. f you want to contest any of these choices or numbers, please write and make the case.
Related
Discover more from diacritical
Subscribe to get the latest posts sent to your email.
