The Mining of Media (or The "Streaming Wars" are Just a Battle)
There is a strange disconnect between our collective obsession with media and the amount of money we spend on it. Every weekend, hundreds of headlines are written about the latest box office blockbuster or bomb. We talk non-stop about Miley Cyrus’s latest antics and the feminist bravado of Beyoncé. A faux reality show comprised of C-list celebrities contributed to the election of the 45th President of the United States. Each day, the average American watches more than five and half hours of video, in addition to some two and half hours of music. Many of us insist on filling every available moment with media of some sort, regardless of whether we’re working, driving, cleaning, exercising, lying in the bathtub or trying to sleep.
Despite the intensity, frequency and size of our media appetites, consumers worldwide don’t really spend much money on content. This has persisted even in the digital era in which the adage “consumers will pay for content” has taken hold. The largest category globally is TV, which has only $300B in annual consumer spend. This sum sounds like a lot, but it’s surprisingly low given the supposed scarcity of human attention and intensity of competition for it, not to mention the total volume of TV consumption (6B people watching two and a half hours per day, with the wealthiest 300MM averaging more than five). What’s more, most expect this sum to shrink as the bloated and barely competitive pay-TV ecosystem gives way to lower cost or free substitutes such as Netflix, Disney+, Amazon Prime Video, Apple TV+, and so on. Spend on books and video gaming are only $140B each. The global theatrical box office is only $41B, a surprisingly low sum given its share of the watercooler. Music is an even more modest $33B.
And not only is consumer spend low overall, we’re incredibly resistant to incremental media spend. Everyone hates their pay-TV bill, but $80 per month is an incredible deal for the average family (who will watch 450+ hours monthly). Film ticket prices are up 12% since 2000, but this amounts to only one extra dollar and is no different than the inflation adjusted prices of the late 1980s. In addition, we now get far better experiences for a ticket – such as larger seats, reserve seating, better screens and audio, etc. Still, per capita ticket sales have fallen from five per year in 2000 to less than three and half today. Most people would rather save $10 on a ticket and watch something for free at home.
The music industry is growing faster than it has in decades, but at only $33B in consumer spend, it could double and still feel anemic. $120 for a year of Spotify or Apple Music (i.e. access to 40MM tracks with no ads and unlimited play time) is an incredible value given the average listener spends nearly 90 hours a month on the service. What’s more, growth rates are likely to slow considerably in the coming years as almost all new subscribers are coming in via $60 student or $150 family plans than span up to six users. Since 2013, Spotify ARPU has fallen from $9.61 to $5.61 – a clear reflection of overall willingness to pay. Some believe this fee will grow over time (and it probably will), but even at the music industry’s peak (i.e. the glory days of the CD album), spend peaked at $100 per capita.
Gaming is going to grow a lot in the future. However, much of this value will come from the displacement of video. It is also notable that half of the market is now “free-to-play games” that require $0 upfront (again, we resist spending money) and then use casino-like tricks to generate revenue in “micro” payments. And even then, an average of 96% of users still spend not a penny and 50% of revenue comes from the 0.4% willing to spend a lot. As regulators and consumers fight back against this model (the worst version being so-called “loot boxes”), the growth may also slow. The shift towards Netflix/pay-TV-styled “all you can eat” content subscribers, such as Apple Arcade, Google Play Pass and Xbox Game Pass, may also curb category revenue.
However, the most instructive example of the value-spend disconnect is likely print. We all believe news is important – culturally, politically, educationally – but few people are willing to spend even the smallest sum on a daily paper. The New York Times is the most valuable print news company on earth and it has only 4MM subscribers and 5B enterprise value. Oh, and nearly a quarter of those subscriptions are to NYT’s crosswords and cooking subscriptions, rather than its news service.
The Synergy is the Message
There are numerous hypotheses behind the dynamics listed above, including low barriers to entry and even lower consumer switching costs (and certainly, the internet has lowered both even further). But what’s important is even more academic – if media matters to consumers, but they won’t spend a lot on it, the efficient use of content is actually to drive other industries, categories and products that have better (and bigger) economics. And this is the dominant model we’re seeing today: [See carousel for descriptions]
Synergies Killed the SVOD Star
There are, of course, slight variances in the models of each of these SVOD services. Some of them charge, some don’t; some will generate direct profits, some won’t; some aspire to maximize view time, others just want a part of it. However, they all share a common business strategy: they’re not focused on the video TAM (“total addressable market” value). Instead, they’re being used to access other, more lucrative ones. Sometimes these TAMs are based around high-margin consumer devices, other times it’s trillions of dollars in retail spend, hundreds of billions in wireless service revenue, or terabytes of consumer data.
For the first eighty years of television, a company launched a video network to collect a share of the video pie. Apple, Amazon, AT&T and Hulu are using their networks to then sell other networks and in doing so, collect a commission against the TAM they don’t personally operate.
Even advertising reflects a desire to access the largest TAM of all: the overall economy. For more than a century, the advertising industry has been a stable 1.25% of total GDP. In 2020, that will be worth $200B. Here, too, AT&T’s plans for HBO are telling. In 2021, the company will launch an ad-supported version of HBO Max. Even the best network on earth hits a low ceiling.
It's also worth noting that this change isn’t unique to video, either. Amazon, Apple and Google each operate structurally unprofitable music services because the money they lose per subscriber is outweighed by the extra revenue generated elsewhere. In 2017, legendary music producer and Apple Music executive Jimmy Iovine even boasted to Billboard that “[independent] streaming services have a bad situation, there’s no margins, they’re not making any money… Amazon sells Prime; Apple sells telephones and iPads; Spotify, they’re going to have to figure out a way to get that audience to buy something else.” To point, Spotify, the only at-scale pureplay music service and the overall market leader, has a 25% gross margin (and royalty agreements cap margins at 30%) and a negative 2% profit margin. Amazon Music Unlimited, meanwhile, gives Prime customers a 20% discount – even though Amazon’s per customer costs are even higher than Spotify’s. Amazon and Google are moving into video gaming primarily to benefit their cloud computing businesses. Microsoft’s Mixer is now spending more on talent than the market has ever seen, but this strategy too seems focused on driving Microsoft’s Xbox gaming ecosystem overall, rather than the video game streaming category itself.
Further still, even those without media assets and who typically sell media are now adopting the model. Instead of selling its subscribers Disney+, Verizon is going to give it away free to reduce churn. And it’ll instead be paying Disney a few dollars per active subscriber in exchange. T-Mobile, meanwhile, does the same with Netflix and has announced opaque plans to offer Quibi, too. Sprint bundles Hulu. Many of these telecom providers also offer music services such as Spotify, Pandora, and TIDAL free, too.
Going Back
Amusingly, the shift to using video to sell other things reflects a reversion to where TV began. In the broadcast era, networks didn’t charge consumers to watch their content. Instead, they were in the business of charging advertisers for the right to their viewer’s attention. We spent a while shifting away from this model – to the point in which companies focused exclusively on consumer spend, such as HBO, emerged. However, this model slowly started to return as pay-TV was discounted to drive broadband and/or landline phones, with deep discounts applied to HBO, Showtime and Starz in order to increase pay-TV customer retention.
Crucially, the strategy of giving media away for other purposes was historically limited by a core problem: it was expensive to distribute content. Cable television, for example, required the coaxial cable to be buried across the country and regularly repaired and installed. Releasing a feature film meant playing it in theater, which meant people had to be hired and physical real estate used. Selling music and video games, meanwhile, meant manufacturing physical media, shipping physical media, and having someone sell it in a physical store.
Today, however, the marginal cost of distributing content to an incremental consumer is approximately zero. And the consequence here is profound. Under this model, the unit economics of all viewers/subscribers improve as you add more viewers/subscribers (e.g. a $100MM show costs $100MM irrespective of how many users you have, but the per user cost goes down). This doesn’t mean giving away said service for free or at an artificially low price solve for profitability. However, it does mean that the more people who have your service, the less you require from each subscriber to generate a profit. And if you have a guaranteed business model – say, selling another iPhone, Showtime subscription, ad impression, datapoint, or two-year wireless subscriptions – there are incentives to maximize your userbase (more people you can upsell to) and the amortization of fixed costs (the higher your unit contribution).
And so, in a weird way, Netflix – which has spent a decade teaching legacy media how to survive in the digital era – may have the least progressive business model. Netflix is just in the $300B business of selling content. It even turns down the $200B+ market for video advertising! This is funny. For years, the dominant bear thesis for Netflix was that every big media company would launch at Netflix-style service, only with “better” content and a deeper balance sheet. Turns out no one will.
Ultimately, this may mean the “Streaming Wars” never happen – or not directly, at least. When the future of the world looked like picking from two dozen Netflixes, it made sense the average household would settle on only a few. But most consumers won’t need to make that choice. At least 70MM US homes have Amazon Prime, so they get Prime Video. Today, 200MM iOS devices are sold per year in the United States, so at least 75MM homes will get Apple TV+ Originals (an Apple Prime bundle will grow this even higher). Tens of millions of AT&T households will get HBO Max for free, as will tens of millions of Verizon Wireless customers receive Disney+ or Quibi, with millions of Disney theme park customers getting Disney+ free, while Comcast customers will get Peacock, and so on. The “Streaming Wars” are really just a single battle in the “ecosystem wars”, which, in most cases, is fought asymmetrically. Unlike Amazon, Apple isn’t an e-retailer, nor is it an enterprise cloud services provider. But the two companies offer similar video services with similar monetization models. Netflix is in the crossfire of this ecosystem war and will need to excel through specialization and a first-mover advantage to be sustainable. But for the rest, the “war” is mostly won through synergies and monetization elsewhere. It’s not about maximizing the share of a consumer’s video time or video spend or video data, but monopolizing a consumer’s free time, spend, and data.
American Dolls
The conglomeration and strategic deployment of media has been in process. As an example, consider the film industry. Many consumers hate the idea that film studios seem to exist only to produce merchandise, sequels and branding partnerships – but this trend is the inevitable result of a history of thin margins, shallow balance sheets, and small, unpredictable profits. In the 1970s, Fox’s $10MM loss ($70MM in 2019 dollars) on Hello, Dolly! – a film which starred Barbara Streisand at her peak and won three Academy Awards – forced it into insolvency. Once the studio had repaired its finances, it diversified its way into resorts, soft drink bottling and film theaters. After its own bout with disaster, Columbia Pictures was purchased by Coca-Cola (!!!), which later sold the division to TV maker Sony. Today, every studio is part of a much larger enterprise. Not only does this mean it can withstand failure, but its parent can now recoup this investment across everything from theme parks, consumer products, broadcast and cable television networks, and consumer hardware and/or telecoms service. And given media can now be delivered at no marginal cost, it makes sense to deprioritize its revenue model for other areas.
It’s true that many media businesses achieve decades of high margins and reliable revenues. However, this generally reflected high barriers to entry and the regularity of consumer need, rather than the value of content itself. Once it was possible to distribute content everywhere and at no cost, for example, the print industry’s subscription and advertising revenues imploded. As Benedict Evans has said, companies like NYT were “actually [in the] trucking & light manufacturing businesses”, something few companies could do let alone quickly replicate. The reduction in other pre-digital advantages, such as foreign bureaus, further eroded margins. Pay-TV is the best example. Not only was competition capped for decades (first by finite broadcast spectrum, then by as little as what channel number a new network would get), but every player succeeded together (hence the bundle), no one network could satisfy all audiences (you could only air one thing at a time), revenues were guaranteed (via multi-year distribution agreements + clear advertiser model as long as viewers showed up), and customers could never be lost (they were always a channel surf away). That’s a good business if you can get it. But as OTT distribution has deflated almost all of these advantages, it’s no surprise that margin profiles and ambitions have changed. (Notably: this doesn’t mean there can’t be direct profits in SVOD. As competition cools and winners become clear, prices will go up and each conglomerate will look to optimize the performance in the BU.)
This brings us to the final, if anti-climactic question: what became of the long-foretold battle between Hollywood and tech? It essentially deflated. The big tech companies never tried to kill the media industry, nor did it really invade it. Meanwhile, the media companies mostly sold themselves. NBCUniversal and Sky exited to Comcast, a telecom, while Time Warner sold to AT&T, a telecom. And both of these telecoms now position themselves as “tech” companies and are investing in large digital advertising and data arms, as well as multi-product/service platforms. Another giant, Fox, closed up shop rather than fight the digital-era war. And while it sold to Disney, it’d be imprudent to call The House of Mouse a “content” company. It’s powered by and excels at content, but 34% of its revenues come from theme parks (which have a net asset value of $60B), another 8% from consumer products, and 15% from the resale of sports rights it doesn’t directly operate. A few pureplay media companies exist, but few are optimistic about their future beyond supplying content to these tech or tech-telco companies. And the tech companies themselves have built businesses around selling other companies’ media subscriptions – they certainly don’t want to destroy them. At least, not yet.
Matthew Ball
