Introduction
For those who grew up on the internet in the early-to-mid 2000s, “digital piracy” is practically synonymous with music. Napster, LimeWire, and the RIAA suing teenagers were among the defining battles of the early consumer web.
Given that, if you jumped two decades forward from then and looked at the state of piracy today, you might be surprised by the headline stats – MUSO (a firm that tracks piracy by mediatype)1 indicates that interest in pirating music is about 10% that of television (8.4 billion visits vs 83.4 billion), and film piracy visits add another ~16.8 billion visits. And lest the large dropoff in 2025’s overall traffic numbers make you think that this is some kind of outlier, the 2024 report backs up the substance (even if not the specific round 1:10 ratio), with music piracy a fraction of television piracy and (apparent) interest in music piracy decreasing faster than in television/film. The comprehensive 2017-2023 review of piracy in the EU showed the same trend over a longer timescale2.
If, however, you subscribe to what I’ve come to call Gabe Newell’s Axiom of Friction3:
Piracy is almost always a service problem and not a pricing problem […] If a pirate offers a product anywhere in the world, 24 x 7, purchasable from the convenience of your personal computer, and the legal provider says the product is region-locked, will come to your country 3 months after the US release, and can only be purchased at a brick and mortar store, then the pirate’s service is more valuable.
… this will seem like the obvious outcome.
After all, no matter which music streaming service you pick (e.g. Spotify/Tidal/Amazon Music/Apple Music), you’re unlikely to worry that your favorite songs aren’t going to be available. You pay one price, you get one service, and you pick your service based on how well it integrates with your existing flows/preferences, or what quality options (e.g. lossless audio on Tidal) are available, or UI/UX, etc.
By contrast, streaming video is a menagerie of digital fiefdoms, where you’re charged a toll by specific entities based on what, specifically, you want to watch. Stranger Things? Gotta pay Netflix. Star Wars? Disney wants to talk. The Office (US)? Peacock wants to know your location. And I do mean “wants to know your location”, because heaven forfend you want to watch something made by a company in a country that’s not the same as where you’re currently located. The gods of licensing laugh at the hubris of those who try, and drop them into a labyrinth of geographical exclusivity and/or localization rights4.
That difference is well-known and much-bemoaned. What’s interesting to me is how that difference came to be. After all, it’s not like the film industry didn’t launch an equally rabid assault on digital piracy in the 2000s – who can forget the legendary “You Wouldn’t Steal A Car” campaign – and yet the respective industries’ piracy numbers have diverged in the way we see. Both wanted to be digital landlords in response to the exogenous shock of digitization, and both tried to lock their catalogs in exclusive storefronts, but only one succeeded; I posit that’s because record labels never owned the record stores.
How We Got Here
You might be wondering what I mean by that, or why it’s even relevant. A bit of pre-digital business history is necessary context for understanding that.
If delivering content directly to consumers was so lucrative that everyone wants to get in on it now, why wasn’t it always that way? Why didn’t you buy VHS tapes and DVDs directly from Disney? Why was video rental its own separate thing?
In the world where art was transmitted by moving atoms, there were two reasons:
There is a vast difference in operational and capital requirements between “creating a song/film” and “running thousands of brick-and-mortar stores and handling sales”. Being good at one has no bearing on whether you’re even capable, never mind competent, at the other.
Even if labels/studios wanted to vertically integrate into retail, the right-of-first-sale5 meant that, once you actually sold a thing, you could no longer control who resold or rented it6. You legally couldn’t prevent anyone from strolling in, buying copies of everything you sold, and then setting up shop right next door/across the street and reselling or renting out those exact copies. Exclusivity couldn’t be meaningfully enforced.
Copyright and The Music Industry
When Congress passed the Copyright Act of 1909, they included in it a compulsory licensing rule for musical compositions (note: not recordings, just the notation/lyrics), which meant that (in exchange for the US government using its coercive power to enforce the copyright) music composers couldn’t prevent anyone from covering their songs, so long as they paid the required royalties. Sound recordings of songs had no standard copyright protections at all until the 1970s, when the international Geneva Phonograms Convention was… uh… convened7. This meant that labels had no legal means to prevent people from making unauthorized copies of their work or replaying it for others… which might not seem like an issue until you remember that radios existed. Every single radio station was allowed to play and transmit songs over AM/FM radio for zero marginal cost per-play; they needed only to buy the original record for the song and hook up a setup to transmit those sounds via radio8.
The music industry rolled with the setup. They treated radioplay of their songs as advertising. After all, if you wanted to be able to hear their songs whenever you wanted (as opposed to whenever some dude in a radio station decided), you’d have to buy their record. Radios were treated as a discoverability vector, in the same way that no self-interested TikTok creator would demand a fee for putting their shorts on your FYP. Combine that with jukeboxes, where customers paid to select a particular record/song for play, and you had an industry where presence was ubiquitous, full control over reproduction was never a given, and good business relied on getting people to come back and listen to your ~3 minute tracks repeatedly (either because they bought the record or because they paid for each play on a jukebox). Moreover, jukeboxes made more revenue when they could meet the desires of more listeners, which meant that rather than having any given jukebox be dedicated to a particular genre/label, the owner/operator was incentivized to keep a variety of options, and to regularly swap out those options based on changing tastes or just for freshness.
Copyright and The Film Industry
By contrast, Hollywood has basically always gated access. Unlike record labels, Hollywood studios actually did own the “stores” (the movie theaters) until the Supreme Court found, in United States v. Paramount Pictures in 1948, that this was a violation of antitrust, leading to the (recently revoked) Paramount Decrees from the DoJ, enjoining studios from owning theater chains. Even afterward, video consumption remained gated, windowed access rather than the ownership model of records. You’d pay for a ticket, or tune into a specific channel, and view the experience on their terms. You paid them for access, not ownership. And when the FCC repealed the Financial Interest and Syndication rules in 1995 (which was the functional equivalent of the Paramount decrees for television channels), studios actually were allowed to own the stores, or at least some kinds of stores, which led soon after to the merger of ABC and Disney (among others).
In short, the concept of “ownership” was core to the music industry from its early days, on both the consumer and producer sides, and the industry structurally evolved to reflect that reality. Until the advent of VHS, though, the film industry had no similar concept of “consumer ownership” – the IP owners controlled (or licensed) when, where, and how you were allowed to experience their IP.
Rejection, Or When The Music Industry Tried Landlordism
Then came computers and the Internet, which blew up both constraints of the physical world simultaneously:
“Distributing art” no longer required physical locations; and
Because digital files can’t be handed over without copying, they were licensed rather than sold, and right-of-first-sale no longer applied.
In short, the limitations that the music industry had been forced to work around for nigh on a century had just gone up in smoke, so even as it made piracy easier, it also opened up a brave new world of monetization-via-rent-seeking.
Round 1
The industry made an early effort: in response to the deluge of downloads of their tracks (for which they weren’t being paid), they launched attempts to capitalize on digital distribution with newfangled “streaming services”. Let’s review the offerings (at least, as best as we can a quarter-century after they were announced), shall we?
In this corner, PressPlay (stylized as pressplay, originally Duet):
Team-up of Sony Music and UMG (hence “Duet”)
$15/mo for the lowest tier (in 2001, mind – inflation adjusted, ~$28/mo today)
50 downloads and and 500 onetime streams per month (or at least, so I extrapolate by synthesizing the description given here for the $20/mo plan with the above reference for the entry price/numbers)
Eventually, an announcement that you could also burn some of those downloaded songs to CDs!... As long as you went exclusively through Roxio, of course. Oh, and as that earlier article noted, only 10 songs/mo.
Eh? Not bad, right? OK, yes, it’s obviously really bad by modern standards, but this was 2001, before venture capitalists were willing to let you burn billions of dollars without any semblance of profit in a fight for market share; hidebound shareholders were still concerned with outdated concepts like “return on invested capital” and “net profits”.
At least, that’s what you might say if you were a music executive trying to defend this venture in retrospect. You’d probably also try to sweep uncomfortable facts like “not all songs available to download” or “you can only burn one song per artist per CD” under the rug.
OK, so that one was a bust. What about the competitor?
In the other corner, we have MusicNet… a venture that was apparently so disastrous that, despite its mention on PressPlay’s Wikipedia page as “RealNetworks’ competitor”, it doesn’t even warrant a name drop on RealNetworks’ own Wikipedia page, instead being obliquely referenced in a sentence as “the first coordinated effort to sell and deliver music from major record labels over the Internet”, which if you didn’t know what to look for might not immediately connect. Luckily, the LA Times digitized their coverage of its launch, so we know (at least one version of) what was on offer:
Catalogs from AOL-Time Warner, EMI, BMG (the owners/investors) and Zomba (no, I don’t know what that is either)
$10/mo (~$18/mo today)
100 (presumably one-time?) streams and 100 downloads
Those downloads can only be used on the machine that originally pulled them from the service, though – no CDs, no Walkmans, etc.
The downloads also expire after 30 days, so you can only ever have 100 songs in your digital library
Yeah, probably easy to see why this also foundered.
The biggest note, regarding both of those attempts, is that (despite the LA Times’ implication in their piece to the contrary) even for participating labels, you wouldn’t find their entire catalogs on the services because artists balked at receiving less than a quarter of a cent per stream/download… after deductions for “CD packaging costs”, of course.9
Hm.
Anyway, then Steve Jobs came along and sold the labels on the radical idea of “maybe just let consumers buy and listen to music”, and the rest is now ancient history.
Or at least, the narrative might seem very neat and tidy that way, but in reality, label executives weren’t that blind/ignorant. When PressPlay and MusicNet were initially announced, European antitrust investigators and the US DoJ opened investigations about whether this might be cartel-esque behavior, and both (not-yet-existent) services hurried to clarify that “licenses from their label backers are nonexclusive and that they hope to cross-license so each service will have a broad range of music”, even as representatives in Congress were drafting bipartisan legislation to force labels to license their music at the same terms for everyone. Even before anything more than the names and participating partners of these ventures was known, (some) analysts were calling the shape of the eventual solution:
“None of this is going to work if all five of the majors aren’t involved,” said David Bench, an analyst at Arnold & S. Bleichroeder. “Users want to go to one site, get all their music – cheaply. That’s the bottom line.”
There were plenty of signs that exclusivity wouldn’t work… but when you’re the owner of the record store and also the creator of the records, and the same is true for your competitors, cross-licensing is voluntarily disarming your nukes without proof that the other side will do the same without pulling any funny business. The labels needed a third party (i.e. Jobs) because nobody could trust anyone else to be the “neutral” seller.
Round 2
“OK, maybe we can’t outright own our own record stores that exclusively sell access to our catalog”, the labels said, “but what if we picked which stores had dibs?”
In the days where physical distribution was king, this was a losing proposition because you were effectively cutting off your nose to spite your face.
What, you don’t believe me that this was a dumb idea? OK, fair, because folks in the music industry also didn’t until they got burned. Let’s look at one (in my opinion particularly puzzling) example:
Alanis Morissette records an acoustic version of “Jagged Little Pill” for release alongside the album’s 10th anniversary.
Warner Music (through Maverick Records) makes the decision to exclusively sell the album through Starbucks in North America for the first 6 weeks after release.
HMV retaliates by pulling all of Alanis Morissette’s discography from their stores (in her native Canada!) for that entire exclusivity period.
Oops.
“OK, so we don’t want to ruin our relationships with the stores we rely on to distribute our physical product, but in the age of primarily digital distribution, what if we just had exclusive streaming partners?” they asked. “If you want to stream it, you need to go to one particular place. Surely that’ll be fine?”
It wasn’t.
Kanye West, in 2016, announced with great pomp that his album The Life of Pablo would be exclusive to Tidal. Tidal promptly topped App Store downloads. What this meant in practice was that if you wanted any of the songs in your playlist, you either needed a Tidal subscription and your entire playlist shifted onto it forever (and pray nobody else had the same idea with a different service), or you had to pirate it. Pretend to be surprised which one people picked:
Regardless of how the album is finally released, more than 500,000 people have already downloaded pirated versions of “The Life of Pablo” since it popped up on Tidal, according to TorrentFreak.com.
Later that year, after Frank Ocean used an Apple Music exclusive to very publicly walk out of his contract with them, UMG made streaming-service exclusive arrangements verboten, and major-label streaming exclusives have been functionally extinct since.
Round 2.5 – How About Podcasts?
For those making the argument that music isn’t really comparable because you listen to songs more than one time and they’re short (i.e. replayability and velocity), I hear you and you’re not entirely wrong. However, that argument still has to contend with Spotify’s failed attempt to fence off their own podcast fiefdom at the start of the pandemic (though I don’t think that was deliberate timing). It might be the inverse direction (i.e. the record store trying to buy the labels), but the end state is the same.
To keep this recap short, starting in late 2019/early 2020, Spotify bought or exclusively signed a slate of podcasts (most notably The Joe Rogan Experience and Barack/Michelle Obama’s shows from “Higher Ground”, their production company). However, this push didn’t last – listener numbers cratered, creators wanted out (the Obamas as early as 2022), and a billion dollars later, Spotify cut its losses on the experiment and let the big names distribute freely again while the ones that were too far gone became in-house staff or shut down entirely.10 Why? Because podcasts had grown and developed in the RSS ecosystem, where the promise was that any player could play any podcast, just as music lovers came to expect friction-free, ubiquitous music thanks to a century of standardized records, CDs, and players.
The “Golden Age of Streaming” and The Great Clawback
If Hollywood, by contrast, had been built on gated access and studio-owned channels, why did we get what felt like a decade-long Golden Age of Streaming, where $8/mo bought access to basically all of television? Because video streaming bootstrapped itself out of the one corner where first-sale doctrine actually applied – video rental.
When Netflix was “just” a rent-DVDs-by-mail service, they were covered under right-of-first-sale; they didn’t need studio permission to send you a copy of, say, Speed or The Princess Bride so long as they were able to get their hands on the media.11
The Golden Age
When they moved into streaming, though, they did need to get licensing permission; after all, they needed to make copies of the videos that they were sending over the wire to customers, which is the exact thing copyright requires a license to do. Studios initially gave Netflix these licenses cheaply and without many strings – “it’s just renting but without the DVDs!”12 was essentially the reasoning, combined with existing relationships between studios and Netflix (see footnote 12). Because studios weren’t stingy with licenses, Netflix could pass those low costs on to consumers in the form of e.g. low subscription costs, large catalogs, and reinvestment in a better service.
However, the 2000s gave way to the 2010s, and the pressure of capital markets incentivized finding ways to grow otherwise stagnant revenues/profits.13 Licensing fees for digital streaming made a lot of sense to look at first. Moreover, Netflix was no longer the only player in the space – NBCUniversal and News Corp (through 20th/21st Century Fox) had teamed up to create Hulu (later joined by Disney), Amazon had launched its own Instant Video service packaged with Prime Membership, and the proliferation of other streaming services (both legal and otherwise) enabled by the lower cost/complexity of setup through e.g. cloud service offerings made the path of returning to the era of fully-studio-owned media distribution a feasible reality.
This was further incentivized by Netflix’s own entrance into original, exclusive content. Offerings like House of Cards and Stranger Things, as well as HBO’s Game of Thrones demonstrated that users would be willing to subscribe to streaming services if one or more marquee offerings were otherwise unavailable. As physical distribution continued to decline, and Netflix’s subscriber numbers continued to grow, the path forward was both obvious and incentivized.
The First Cracks
The earliest indications of this came from Starz, who (seeing Netflix’s success) wanted a much larger slice of the pie. They’d originally given Netflix the rights for $30 million/year, but by 2011 they’d realized that was a mistake, and decided the threat Netflix posed to their own business model warranted a tenfold increase in the licensing costs. Back then, Netflix called that an “exorbitant” increase, but in retrospect, it seems like Starz likely was closer to the true value. By 2016, Starz had spun up their own streaming service for the content that they used to license to Netflix.
Viacom moved its own programming from Netflix to Prime Video in 2013, widely credited to Amazon being willing to outbid Netflix. In 2015, Epix, the premium subscription cable channel for Lionsgate, Paramount, and MGM, pulled its own content from Netflix in favor of Hulu, an announcement that tellingly prompted Netflix’s then-head of content (now co-CEO) Ted Sarandos to explicitly point out that Netflix was working on exclusives specifically to prevent this kind of thing being more of an issue in the future:
It will take us time to build a robust slate of original movies, but we’re hard at work on it with such great stars and directors as Brad Pitt, Ricky Gervais, Judd Apatow, Angelina Jolie, Sofia Coppola and Adam Sandler.
(from a now-disappeared Netflix blog post, but archived)
It’s not surprising that they felt the need to proactively talk about exclusives; after all, just the prior year, CBS launched All Access as the (best as I can find) the first fully-first-party exclusive streaming service. In a similar vein, 20th/21st Century Fox had also pulled their own catalog to be served up on Hulu (which, as a reminder, was a project they’d started with NBCU).
The Floodgates Open
In 2017, Disney announced they’d be pulling their content from Netflix explicitly to set up a competing service (slated to launch in 2019, so their content would stay through the end of 2018). That, seemingly, was the shot that indicated to everyone else that they should head for the exits from the Netflix party – 2019 saw announcements that The Office (US) would be leaving for what would eventually become Peacock, followed shortly by Friends being pulled for HBO Max, and the freshly-merged ViacomCBS rebranded CBS All Access as Paramount+, your exclusive streaming home for all of their content, in 2021.
In short, all it took was a decade to go from “sure, you can stream our stuff”, to “oh wait, we want more money for you to stream our stuff”, to “actually we’re taking our ball and going home”. Sure, there was some grumbling – who wouldn’t when they had to pay more – but, as the ViacomCBS CEO Bob Bakish said at the time:
All the data suggests, and we’re still relatively early in the streaming world, that overwhelmingly people will subscribe to more than one service.
I mean, I guess he wasn’t exactly wrong. Maybe he just ignored the part of the data saying other, significantly less encouraging things… or maybe they were never aware of those parts? Maybe this is all one big misunderstanding! Surely if they just knew that pirates are also in the paying audience, and that oftentimes pirates spend more money on legitimate access than non-pirates, then we could go back to a beautiful world of unified streaming catalogs that compete on quality and features!
The Other Sides of The Argument
To be clear, this is just one reading of business/industry history, and it’s not bulletproof. There are some confounding factors worth noting:
(Some) record labels took ownership shares in Spotify – One could argue that labels essentially do own the record store as a result. It’s not altogether illegitimate; the argument can be (and I believe has been) made that UMG’s ban of streaming service exclusives was in part to protect the value of their own stake in Spotify.
This may be true. However, given that Spotify is a publicly traded company, finding a list of “largest shareholders” isn’t exactly difficult, and you’re not going to see any familiar record labels on that list.14
Now, the version of this argument that might hold some water is that labels had enough of a vested interest prior to the IPO that, even if they’ve since divested/taken profits, they were incentivized to allow/support Spotify’s success. That’s more compelling, but then one would ask why e.g. UMG would ever even bother with/approve “Apple Music exclusives” if they had enough of a vested interest in the success of Spotify to modify their market behavior.
That said, UMG’s 2016 ban may indeed have been about protecting their stake, but if it was, it’s because they recognized that Spotify’s value comes specifically from serving as “neutral infrastructure” for music.The Production Cost of Music vs Movies/TV is significantly different – This is true, but on its own it doesn’t really explain the difference in industry structure or consumer piracy trends. It explains why studios have a greater incentive towards rent-seeking behavior, but that can’t be extrapolated to “and therefore they must each have their own digital fiefdoms” – studios negotiated perfectly functional revenue-sharing agreements with Blockbuster and (most) other rental services (including Netflix, before streaming). That, then, seems like the anomaly (along with the brief Golden Age of Streaming), one forced into existence via right-of-first-sale and Congress’s refusal to give video rentals the same treatment extended to record rentals.
Music streaming sucks too – Specifically in the context of how much artists are paid (or perhaps more to the point, aren’t paid).
This one’s just true; nothing much to be said or done about it. Unified catalogs help address the problem of piracy, and in so doing maximize legal revenue flowing from customers to producers. It can’t on its own do anything about how that revenue is then divvied up.This only talks about the USA – Guilty as charged. The piracy stats I presented are worldwide, but the mechanism(s) I’ve considered here are almost entirely about US law; other countries may have followed their own paths to similar destinations.
I could try to pretend that the hilariously outsized impact of American influence on the structure and functioning of other societies means that the American story is the tail that wags the dog on this… but the real answer is “I only realized this problem close to the time I finished writing most of this, and I’m too lazy (or perhaps not sufficiently insane) to try to rewrite it to fit the shape and structure of all national jurisdictions/industries”.
Conclusion?
So, 4000+ words later, where does that leave us?
In short, two industries were hit by the same challenge and opportunity (digitization and ensuing piracy) at approximately the same time, but wound up on opposite sides of Newell’s Axiom of Friction.
A century of radio, jukeboxes, and any-record-any-player trained music listeners to expect unified libraries and ubiquitous access. Each attempt by labels to turn catalogs into exclusive company stores or walled gardens in their shift to delivering digital files triggered some combination of customer, artist, or retailer outcry, oftentimes all of them at once. Eventually, labels accepted that a neutral catalog was the only approach that didn’t bleed customers.
Hollywood, by contrast, spent nearly a century training audiences on gated theatrical windows and studio-affiliated channels. The “Golden Age of Streaming” wasn’t a new tech-enabled, stable industry equilibrium; it was a happy accident, a temporary pricing error born from Netflix’s first-sale-enabled DVD rental business. When studios realized they had accidentally recreated the neutral record store for pennies on the dollar, they spent a decade clawing their catalogs back, and the market tolerated that fragmentation because flipping between channels (read: landlords) is how television has always worked… even as enough people seemingly balked at the friction of five separate monthly tolls to keep video piracy sitting at roughly ten times the traffic of music.
Of course, “winning” the battle for a unified catalog in music only meant that consumers won friction-free access. It says nothing about actual ownership of these things we pay for. You may have noticed that, unlike prior pieces, I didn’t include an anecdote of how I fell down the rabbit hole for this one yet. And that’s because this entire piece is, itself, a rabbit hole I fell down while investigating a different rabbit hole.
Maybe I’ll write that one up next. If, y’know, I ever manage to climb out.
\0
Caveats on MUSO data – they track piracy page visits by media type, not “units of pirated content” or “volume of pirated content”. A single visit might be someone visiting a single movie’s listing, a music album’s, or the entirety of a finished-airing TV show. It’s (probably) a good proxy for understanding and tracking interest, but anything more than that is tenuous. The biggest (non-obvious) caveat worth noting for this is that this won’t capture traffic from automated self-hosted setups for media (e.g. things like radarr/sonarr).
This consolidated report cites MUSO numbers as the primary source from which graphs and analyses were generated, so the same caveats regarding methodology/applicability are true here; it also means that the larger trend from 2017 can essentially be extended with the 2024/2025 numbers I linked.
Returning readers might recognize this one from my first post from… almost a year ago?!? When did that happen?
No, I definitely didn’t write the outline for this post after realizing I can watch clips from Utopia (the Australian one, not the UK one) on YouTube all day long, but if I want any continuous narratives I’m doomed to needing to fly to a snake-and-scorpion infested desert continent.
This is more popularly known as “first-sale doctrine”, but I prefer the explicit naming of the right.
Strictly speaking, this was actually reined in for the specific case of music records in the 1980s, when people started renting records and using cassette decks to record them for unlimited later playback, which is just a form of old-school piracy, complete with its own widely-mocked industry PR campaign against the practice.
The USA implemented a for-most-purposes-functionally-equivalent version of the Geneva Phonograms Convention in the Sound Recording Amendment to the Copyright Act of 1909; prior to that, some states had their own versions, and otherwise you had to file suit under common-law unfair-competition.
Even when sound recordings did receive copyright protections in the 70s, to prevent the shape of the radio industry from being upended as a result, Congress didn’t give sound recordings the same public performance right that was given to the underlying compositions, so radio stations still didn’t (and to this day don’t) need to pay the labels (though they do need to pay flat licensing fees for public performance of the underlying compositions).
If it seems odd that artists didn’t want their songs on these platforms, it might help to put the math in perspective: Every month, you (the customer) would pay PressPlay e.g. $19.95 for the Gold tier. That would buy you access to, at most, 825 “song credits” (75 downloads + 750 one-time streams). At a quarter of a cent per play/download, the maximum possible “cost of goods sold” per customer per month is a grand total of… $2.07, if you’re generous and round the partial cent up. The other $17.88 would go to all the other players involved in this, i.e. the music labels who negotiated this amazing deal and the company that built out the service… which is also owned (at least in part, and in PressPlay’s case entirely) by the participating labels.
That’s not to say the experiment had no material results, though – Spotify got a foothold in the podcast advertising/affiliate marketing space and grew that into its own advertising network there, the Spotify Audience Network(SPAN), which is its entrance into the growing “creator economy”.
This is not to imply that studios didn’t try to kill video rental at first. You can find a short history at this VHS Collector article, but in short, video rental was viewed as an existential threat to their revenues/“access control” business model.
You might think that this was viewed as “new income”, as I did when I first started looking into this, but that turns out not to be true. Physical rental stores negotiated revenue share deals with studios that allowed them to acquire stock at much lower costs in exchange for providing some fraction of revenue associated with those titles to the studios, and Netflix negotiated a similar kind of deal as adapted for their subscription model.
For a deeper look at first sale doctrine, its relevance to the film industry, and the effects of digitization on the market from a legal lens, I found this legal note from 2012 quite insightful. One caveat, though: the author takes some seemingly opinionated assumptions about the industry, e.g. Netflix/Redbox labeled “market distortions” rather than just evolution of those markets as technology develops, without any clear justification for that positioning.
That said, labels may still hold stakes that fall below the required-disclosure limits… but at that point, can they really be said to “own the record store”?
