Streaming Music Systems Performance (Infographics)

Two interesting pieces recently.  One on the relative performance of top music streaming options, the other on how those services compensate performers and writers.

Researchers at YouGov BrandIndex looked at a variety of metrics for the top 5 music streaming services in the U.S. They found Pandora to be the dominant player in the field, although Spotify has been making inroads recently.  Pandora has dominant leads in most of the metrics, from number of subscribers to awareness (from both ads and word of mouth).  Spotify's numbers were improving, but the researchers concluded that

"Perhaps the brands with the biggest challenge are iHeartRadio and iTunes Radio. They have reasonably high awareness levels, but do not seem to be getting traction with consumers. The conclusion is that these brands may need to try something different to generate excitement with consumers."

Music streaming services largely emerged as a result of major record companies eagerness to open up a second revenue stream to help cope with declining sales of physical recordings.  Initially, they were eager to license their recordings to streaming services, but faced an initial roadblock - the existing royalty systems employed two distinct approaches.  Royalties for sales were based on fixed compensation for each unit sold, while royalties for licensing music to radio stations was based on a percentage of station revenues (and not directly linked to which music was played).  Conceptually, the radio model seemed closest to how streaming services operated, as well as how audiences used them.  Thus, most of the early deals utilized royalty payments as a percentage of revenues.

As sales in the traditional music markets continued to fade, the record industry wanted more from streamers.  They started arguing that the current system (which they had eagerly negotiated) was "unfair" - largely because streaming revenues were slow to develop.  The attack came on three fronts.
First, that not enough money trickled down to artists and songwriters.  The biggest problem with that argument is the fact that the share that trickles down to the artists and composers is determined by the rights organizations (like ASCAP and BMI) and the actual rights holders (predominantly the record labels), who take their cut off the top.  So the industry argues for a larger royalty rate, of which only a small fraction would actually go to the artists and composers.
Second, streaming services differ from radio stations in that they can and do track individual consumer plays.  There's no mechanism to measure how many listeners hear a song on radio.  The current licensing deal with Spotify calls for royalties to be paid according to a formula that includes both a revenue percentage and the number of streams.  Spotify also pays an additional set of royalties to songwriters and composers for what is termed "streaming mechanical royalties".  As a consequence, Spotify pays a much higher total percentage of its revenues than Pandora.  (Pandora is currently classified as an online radio service, and radio stations are currently not required to pay mechanical royalties).

The third argument is that most streaming services offer a free streaming option, which the music industry argues "cheats" the rights holders because revenues from the ads are less than subscription-based revenues.  The fact that the free/paid proportions for Pandora is roughly 75/25, while Spotify's audience is more of a 50-50 split, also contributes to the difference in royalty payments.  As one record label executive summarized,

"Based on the free model, the payouts we're getting on streaming is so small... The problem that we're running into is Spotify is just not converting users to the paid version quick enough."

That perspective contributed to the fact that the music labels pressured Apple to raise its proposed starting subscription price for the new Beats streaming service (much like the book publishers did for iBook pricing - which the courts later ruled was an antitrust violation).  But the underlying issue is that the record companies want more money, and are using artist payments to engender sympathy.  If artist payments are the real problem, the music industry could solve that easily by granting them a bigger share of the payments they get, or changing accounting practices so that the artist share comes from gross payments, and not what's left after music industry costs (and profits) are covered.

One can look at this situation from the "level playing field" metaphor.  Spotify wants a level playing field by getting the same deal Pandora has, Pandora wants a level playing field with broadcast radio (straight percentage of revenues, and lower percentage), and the music industry wants to raise the height of the field several feet because they cut the grass (i.e. royalties to artists and composers) too short, and aren't making enough profits from their traditional business models.

The current copyright and royalty system is a mess, largely because it was designed to deal with selling physical copies of intellectual property.  The current model has never really worked well with digital reproduction, or with the growing need to replace shrinking sales revenues with licensing arrangements for emerging digital streaming channels.  Add the fact that digital markets are global and have the potential to scale much higher than physical copy sales (tens of millions for hit albums in digital, while in the physical medium heyday, hits sold hundreds of thousands).  Plus, they're now having to deal with younger audiences who care more about access to music than owning copies of music.  In addition, artists need to recognize that the scale differences should be reflected in the setting of royalty fees - and that because digital access to their recordings remain available long after labels drop them, that they'll benefit from their work much longer under digital deals.

The debate and fights over music royalties is likely to continue for a long time, in part because the music industry is trying to hold on to an increasingly problematic business model, and is hoping to find a way to maintain their control over revenues derived from their historic role as the choke point between artists and their audiences.  However, the growth of the digital economy is showing that it doesn't require multiple layers of distributors (and their growing costs) to provide access to products for potential purchasers.  There are already content creators (including musicians) who have discovered that going independent can provide them much higher levels of return, as well as more control over use of their work.  For the big labels, this is a fight for survival; but for society, it's a fight for who gets to control access to content (and who gets to benefit from that).  As for the question of whether streaming will leave artists unhappy - the answer is yes, if the big labels remain in control, and no, if we can shift focus from preserving a declining music industry to how to develop a rights and licensing regime that promotes and protects creation of, and access to, intellectual property.
 
It's time we shifted our concern from protecting the old ways to think about how to develop copyright and licensing systems that benefits the creators and users of intellectual property rather than those who merely reproduce and distribute it.

(For more background, see this post about a digital music licensing panel at the 2014 CES).


Sources: Infographic: Which Streaming Services Are Winning the Battle for Millenial Eardrums,  Adweek
Is the Music Streaming Industry Destined to Leave Artists Unhappy?, Adweek

Copyright Decision on Aereo

Aereo is a recent technology service that offered online access to local TV broadcasts in selected markets in the U.S.  For a fee, they grabbed programs that a subscriber requested, from "free" over-the-air local broadcasters, digitally converted the broadcast into a IP stream that was sent directly to the subscriber's connected device.

And when TV networks and cable systems  heard about it, they sought to pre-emptively ban it.  Cable because it was a much cheaper alternative to a service they sought to provide.  The networks' objection was pithily stated by one network CEO: "If anyone makes money from local broadcasts, we demand a share of it."  A consortium of networks and MSOs filed suit claiming that Aereo violated copyright law and seeking an immediate injunction against deploying the system; however, the judge in that case ruled against an injunction as Aereo offered a reasonable argument that its service was not a violation of copyright - but did not initially rule on the viability of that argument.  Both parties wanted to expedite matters, so they sought to bring the case to the Supreme Court to get a final ruling on the basic copyright issues.

Copyright law is fairly simple, yet complex.  The law gives the owner of the copyright the legal authority to determine the conditions under which the copyrighted material is made available to others.  It also, however, permits a "fair use" exemption under certain general conditions - allowing others to access and use the content under restricted conditions.  The rise of broadcasting created an issue, however - what counts as a "copy"?  This came to a head with the rise of cable systems, and their retransmission of over-the-air broadcast signals from local (and eventually distant) stations.  As a result, Congress amended copyright law to include "public performance" of audiovisual signals, and specifically applied that term to carriage ("retransmission") of over-the-air television broadcasts. 

Around the same time, the Supreme Court issued a ruling on what behaviors were considered "fair use" of audio and video content.  Specifically, they ruled that - for legally acquired content - individuals could record content for their later use (time-shifting), transfer and translate content for use in different locals or on different devices (place-shifting), and make a back-up (archive) copy, but only for their own use.

Aereo's service was designed to fit those "fair use" guidelines and the "free" nature of over-the-air broadcasting (stations are prohibited from charging viewers for access to public broadcasts).  Each Aereo subscriber was assigned their own antenna for receiving the free broadcasts, and unlike cable, content was not retransmitted unless specifically requested by that subscriber, and was made available only to the device the subscriber sent the request from.  Aereo argued that they were not engaging in a public performance, but a private one that essentially amounted to the time-shifting and place-shifting aspects of fair use.

The big media consortium (ABC et al.) argued that the Aereo service was simply a retransmission service, and was engaged in "public performance" because it offered its service to the public at large.  Therefore, Aereo was in violation of U.S. copyright law.

The recent Supreme Court ruling (6-3) was that Aereo's service was "substantially similar" to cable, because it offers a service that allows subscribers to watch TV programs, and that it is a "public performance" because several subscribers may be watching the same program, thus constituting a "public."

It's not terribly surprising, although it is disappointing, that the majority decision is technologically ignorant and focuses on outcomes rather than processes and behaviors.  To the majority, any technology that delivers TV programs to members of the public is essentially no different than a cable system and thus any retransmission right for that content must be granted by the copyright owner (presumably, but not necessarily, for a fee).  In addition, the majority nonsensically assumes that since the copyright act does not explicitly define "public", it is any group of individuals beyond what may be watching on a single device. The majority also finds that it makes no difference who is initiating and controlling the retransmission - an individual subscriber or a cable system; rather, they shift the focus to Aereo because it offers a "service."

There is a weird yet nonsensical example given, where the majority suggests that Aereo's problem is that it doesn't provide a service to the copyright owners, but to the public.  The majority repeatedly focuses on copyright owner's right to set the terms of "performance" or consumption.  This ignores the fact that, under law and the terms of their licenses, local broadcasters must provide their copyrighted content freely and without restriction to the public.  Not only to those watching live at home on a TV set, or through a cable system paying retransmission fees; the viewer, or Aereo subscriber, already has the legal right to view or listen to broadcast content.  The copyright owners have already been paid for that performance by the broadcaster - the majority seems to think that viewers need to pay a second time to engage in time-, place- or device-shifting.

The majority decision argues that this ruling does not prejudge future technologies, saying that it is not their intent.  It's hard, though, to see how this is possible, when the decision sets down three extremely broad definitions - 1) that any system for delivering TV content to consumers is "substantially similar" to cable and thus subject to cable's rules; 2) that any group of individuals is considered to be "the public"; and 3) that the driving purpose is not protection of a copyright owner's rights, but the networks "right" to offer a "public performance" of copyrighted material.  Combine that with the majority decision's total disregard for the specific elements of technology and service offered by Aereo - and thus not providing any hook for a narrow argument - and you set a precedence for overreach.

Further, this case offered the opportunity to re-examine which activities are covered by "fair use."  In light of the wretched quality of the decision, I'm relieved that this Court didn't take up that opportunity.

The dissent shreds the majority decision, arguing that the service provided by Aereo was not even a "performance" as defined by copyright law, much less a "public performance."  Rather than looking at the end stage of the service (providing TV programs to viewers) and making an indefensible leap to equate it with cable, the dissent treats Aereo as an Internet content delivery system.  And supports their argument by actually looking at what the service does, then examining the actual case law relevant to those actions (rather than relying on problematic anecdotes).  The dissent eviscerates the majority's "looks like" argument, noting how it conflicts with other recent decisions (including some authored by the same Justice who wrote the majority decision), and ignores both the question of the type of copyright violation being considered (primary, secondary violation, and whether Aereo is acting as an ISP and thus exempt), as well as the issue of "fair use" and the role of the subscriber (not the system) in selecting content for viewing.

The majority decision has turned a copyright case into a "performance" result, achieved only with mystical inference of Congressional intent, over-broad definitions of "public" & "performance", and a "looks like a duck" equivalence of two video delivery systems that could hardly be more polar opposites in technology and operation.  And by doing so, making the application of copyright to new technologies and content delivery systems even more problematic for the future.

And of course, the TV & cable industry hailed the decision, seeing a potential source of revenues (or a protection of current retransmission rights fees).  However, that's likely to be a short-term and low-value source for broadcasters and networks for two reasons - first, that while they think they're soaking the service operator for these revenues, eventually viewers will figure out that stations and broadcast networks are asking them to pay for "free TV"; second, that it's the content that generates the value for viewers, not their "public performance" of that content.  In fact, to the extent that the broadcast "performance" has any value for the station or network, it comes from the broadcast's ability to reach an audience; as such, any mechanism that will extend or expand their reach should be welcomed, rather than challenged.

Sources -  Supremes Rule Against Aereo, Broadcasting & Cable
ABC et al. v. Aereo, Supreme Court decision No. 13-461

How rampant is online Piracy?

The author of a new study commissioned by NBCU has released a video presenting claims of huge increases in the amount of online piracy over the last two years.  I'm always a bit skeptical of industry-backed research in this area, particularly when the study doesn't provide details on their methodology or definitions, and reports claims of significant numbers in a somewhat dishonest manner.

This report seems to fit that pattern.  Hidden in the passing discussion is the result that their study of online piracy in 2010 reported that 23.8% of all online traffic was pirated content.  Still, while expanding the scope of their study to include new forms of pirated content, and claims that the amount of pirated content distributed via the internet had skyrocketed in absolute terms (160% increase in the amount of "pirate" data traffic), along with an increase in the number of people regularly trading in pirated content (up more than 10% to 317 million unique users), the report still suggests that in 2012, a whopping 24% of all online traffic was pirated content.  Yes, that's right - in relative terms, a nonsignificant increase of 0.2% of total global online data traffic - and that's with a broader definition of "pirated content."

The claim of huge increases in absolute amounts of content, even if accurate, is confounded by the rapid increase in both the number of Internet users in the three areas examined (North America, Europe, and Asia) and the even more significant increase in online traffic driven by growth in video content (and its much larger data files) and the expansion of broadband connectivity.  Put in that context, and looking at share of data traffic, the increase is minimal.  Similarly, a 10% growth in the number of internet users accessing pirated content regularly sounds high, until you compare against the growth in the total number of internet users globally - which is up 17% over just the last year.  The 317 million number also seems a bit suspect when you consider that it amounts to 17% of the internet users in those areas.  Do 1 in 6 internet users really traffic in pirated content at least monthly?

I'm not going to claim that online piracy isn't a problem, or profess any real knowledge of how significant the problem is, or how negative its impact.  (Although I'll point the interested to a study for the WIPO that found that while piracy of broadcasting signals was rampant, the economic impact was slight as the vast bulk of that piracy was in areas where signals weren't being marketed anyway, and/or where populations were unlikely to be able to afford first-world prices).

But I will point out that the trumpeted claims of online piracy becoming a significantly bigger problem in the last few years is contradicted by the report's own numbers, once they are placed in the context of the continued rapid growth of internet use and increased data traffic.

Source - Online piracy of entertainment content keeps soaring, LA Times
Study on the Socioeconomic Dimension of the Unauthorized Use of Signals, Part III - WIPO SCCR/21/2

Fair Use Guide for Journalists

Following up on a project of putting out a set of principles/guidelines for fair use in making documentaries, Pat Aufderheide of American University has facilitated a project to come up with a similar set of principles and guidelines for journalists.  The guide was released a couple of months ago, in connection with a TEDxPoynter talk (video available here).

The final published guide and other supporting material is available here.

PDF version of Set of Principles in Fair Use for Journalism

Is TV Everywhere Legal? For now, maybe...

The last year has seen several new start-up services that seek to provide users access to programs they legally receive at home when they aren't in front of the TV.  And that's part of the goal of TV Everywhere - being able to access and view programming regardless of time, location, or type of screen.

  Aereo is a new start-up that offers subscribers access to their local TV stations through the internet, particularly via mobile devices.  It works by providing subscribers with a small antenna/tuner connected to their home Internet connection portal; allowing subscribers to take their free broadcast TV signals beyond the home.  As soon as the Areo started its service, they were sued by a consortium of networks and broadcasters for copyright infringement.  Part of the suit asked the courts to ban the service while their suit was in litigation - i.e., they wanted to kill the service while the challenge dragged out in court for years.
  An appeals court has now affirmed the trial court ruling against an injunction, allowing Aereo to continue operating through the litigation process.  Normally, such an injunction banning some behavior or service is awarded only if the party asking for the injunction is considered likely to win the case on its own merits.  As such, it's not a clear indication that Areo's service is legal, although judges indicated that Areo had a viable legal precedent for their system falling under "fair use" guidelines (as place-shifting of an otherwise legal signal).  That was enough to suggest that the challenge wasn't a cut-and-dried winner.

While I'm not a lawyer, the economist in me does wonder why broadcasters would object to a service that would make their signals more widely available and more valuable to viewers.  Perhaps it's not TV Everywhere access they object to, but not being offered a cut of Aereo's subscription revenues.

Source -  Appeals court denies broadcaster request to shut Aereo,  Consumer Electronics Net

Using Social Media Images

In what looks to be an interesting case, a Federal judge ruled in a summary judgement that the Washington Post, press agency Agence France-Press (AFP), and Getty Images (which AFP uses to market its photos, and which it used to license the images in question to other media outlets and publications) improperly and illegally used images that a photojournalist had posted on Twitter. 

Daniel Morel, a photojournalist whose photos (of the immediate aftermath of the Haiti earthquake) were placed on the AFP newswire without payment, turned to the courts in March 2010, claiming AFP had violated his copyright. AFP responded with a few different defenses; its boldest, almost shocking defense was that the Twitter terms of service actually granted it a license to use whatever photos it could grab off the micro-blogging service.

 AFP had argued that once posted on Twitter, the images became public - specifically that since the photographer had granted Twitter a license to post and share the picture, that meant that he had granted permission for everybody else to use the photos freely, without permission or royalty.  In this case, the judge ruled that such an argument was dead wrong now, just as it was when it was pointed out to the AFP in a copyright infringement case years earlier.  And it's certainly not an argument that AFP and other media outlets would extend to their own photos and content, or when they sought payment from others for licensing the Morel photographs to other outlets.
  So it was no surprise that the judge knocked down that defense in a summary judgement.  (Judges use summary judgements when the allegations are so silly and/or inappropriate that there's no point in even going to trial).  The judge also ruled in a separate summary judgement to limit potential damages.  Morel's attorneys had argued for damages as set forth in U.S. Copyright Law, which allows for damages of up to $150,000 per instance if infringers did so willfully, and $200 if infringers can prove they didn't know they were infringing; the kicker, though, is that each the damages apply separately to each separate instance of infringement.  That let plaintiff's attorneys to ask for damages for each infringement, for each copy of the photo printed in newspapers, or each time a webpage with the image was accessed online - which could lead to hundreds of millions of dollars in statutory damages).  The judge, in summarily ruling that any damages would be limited to the initial infringement of each image (and not later reuse), essentially indicated that the current legal language on damages was "ridiculous" - at least when applied to online circumstances like this specific case.
  A number of other arguments and aspects of the case, such as whether the infringement was willing, and whether Getty Images acted in the role of purely a middleman passing images and covered by the ISP exemption, will continue on to a full trial, or whether AFP & Getty violated the DMCA by providing false copyright information (in this case claiming ownership by AFP for licensing purposes through Getty).

  I've posted on a similar situation emerging from secondhand use of photos and images on Pinterest.  The main point coming from that situation and this case, is that content creators need to be aware that availability on the net does not equate with being in the public domain.  If you do reuse content, you need to consider the nature of that use, whether permission is likely to be needed, and that proper credit and attribution is provided - particularly if you're commercial operation.
  The one thing I'll disagree with in terms of the judge's summary judgements relates to the damages.  I'll agree that the current copyright law language overdoes damages (mostly to maximize potential deterrence), but rather than limiting damages to the single instance (per image), I'd also argue that Morel should be able to recover a hefty multiple of the revenues that AFP and Getty earned by (illegally) licensing his photographs.  But that's the result of fraud as well as copyright infringement, so I hope Morel's attorneys can still make that case.

Sources - News flash for the media: You can't sell photos grabbed from Twitter,  arstechnica
More than $120,000,000 at stake in AFP vs. Morel case,  British Journal of Photography