Another day, another story of copyright being used for censorship, rather than as an incentive to create. Here's the headline: Gene Kelly's widow is suing to stop an academic book exploring various interviews that were done over the decades with the famed actor/dancer. And here's the lawsuit, in which Kelly's widow, Patricia Ward Kelly, who was married to Gene Kelly for the last seven years of his life, claims that she holds the copyright on every interview that Kelly ever did. From the lawsuit:
The spoken and written words by Gene Kelly during all of his interviews ("Interviews") are original works of authorship and are copyrightable subject matter under the laws of the United States.
Prior to and during Gene Kelly's marriage to Plaintiff, which lasted until his death in 1996, Plaintiff was designated as Mr. Kelly's official biographer and archivist of his materials, including letters, interviews, manuscripts, holograph notes, photographs, memorabilia, and related items. Plaintiff is the sole, official authority entrusted by Gene Kelly to promote and protect his legacy. In these capacities, Plaintiff documented his life and work, and collected, organized and catalogued his materials.
including the Interviews, so that these materials could be used to write books, create online
platforms, and produce films, educational talks and shows, so as to provide an accurate
record of Gene Kelly's life and work.
In accordance with Gene Kelly's Will and the Eugene C. Kelly Family Trust, Plaintiff was bequeathed and succeeded to the rights to Gene Kelly's
intellectual property, including the copyrights in and to the Interviews.
Now, the legal issues here are at least somewhat nuanced. The question of who actually holds the copyright in an interview is actually a hotly debated topic in some copyright circles, and the answer is not as clear or as simple as you might think (or as it probably ought to be). Remember, of course, that the law is pretty explicit that copyright is given to whoever fixes the interview into a tangible medium. So, in most cases, it would seem that whoever is recording/transcribing/publishing the interview likely holds the copyright in it.
That's what a district court in Southern Illinois found in the Taggart v. WMAQ case back in 2000. There, a court found that the interviewer held the copyright, rather than a prison inmate who had been interviewed by the local TV station and didn't like how it came out. The inmate argued that his responses were a "performance" that allowed him to get copyright protections, but the court rightly rejected this:
Plaintiff's reading of copyright law to protect his
interview comments with WMAQ as a work of authorship
conflicts with the “most fundamental axiom of copyright
law [that no] author may copyright his ideas or the facts he
But not all cases have turned out that way. There's a case from 1980 that suggests there might bea copyright interest in the interview that could be held by the interviewee, but the case did not turn on that issue and the court went no further. There's another case that suggests each individual in an interview retains a copyright interest in their portion of the interview (so just the questions or just the answers). And then some argue that the entire interview is a "joint work" of authorship, where both parties hold the copyright jointly. Frankly, I think that copyright law is pretty clear that the Taggart ruling is technically correct, that the ownership goes to whoever does the fixing. But, with weird rulings lately about "performances" who knows how courts will rule.
Frankly, it's a little amazing that the issue hasn't been more widely litigated. But here's a chance to do so, though I suspect it may get tossed pretty quickly, because the lawsuit, at least, doesn't even bother to specify what specific works are being infringed, or even hint at whether or not Kelly registered his copyrights in those interviews (a necessary step to bring a lawsuit). Given those two limitations, the lawsuit, as is, likely doesn't have much of a chance.
The book in question is written by an academic, Kelli Marshall, who appears to be a huge fan of Gene Kelly and is working to put together a scholarly book exploring a bunch of his interviews. Kelly's widow finds this quite upsetting:
On or about March 29, 2016, Plaintiff was contacted by defendant Marshall via a Facebook message inquiring whether permission is needed to include several ofthe Interviews in a printed book Marshall is planning to cause to be issued by and through University Press.
On or about March 29, 2016, Plaintiff responded to Marshall via Facebook message, stating, "Yes, Gene's words are his intellectual property . . . as are his letters, holograph notes, magazine pieces, etc. . . . You must obtain permission to use them."
On or about March 29, 2016, Marshall responded to Plaintiff via Facebook message, informing Plaintiff that Marshall is in the process of editing a book of
Gene Kelly interviews for co-defendant, University Press, as part of the University Press'
"Conversations with Filmmakers" series (the "Book"). Marshall stated that she intends to
use various Gene Kelly interviews, including several interviews Gene Kelly had conducted
with the British Broadcasting Company ("BBC"). Marshall sought Plaintiff's permission for use of those Interviews.
As you can probably figure out, Kelly's widow refused to grant permission, and then followed it up with a cease and desist letter. University Press then sent her a letter saying that it was going ahead with the book, saying that it had obtained permission "from unidentified third parties," which likely means the publications where the interviews were initially published. And you can figure out what happened next:
On April 18, 2016, Plaintiff's counsel responded to University Press
by email (copying Marshall), stating that Plaintiff owns the copyrights to all of the
Interviews, not just interviews with the BBC, and that Defendants have no permission to
use any of the Interviews for the Book or for any other purpose. In that same email,
Plaintiff's counsel advised University Press that the threatened publication is highly damaging to Plaintiff's rights and, unless Defendants cease and desist, Plaintiff would seek damages, including statutory damages, for willful infringement of the copyrights in the Interviews.
Yeah, this seems like a nonstarter. First off, Kelly's widow is clearly overclaiming here. Just because Gene Kelly said stuff in interviews, it does not mean that he has any copyright interest in them, let alone automatically getting copyright on all his words, where no one can ever make use of them. That's just not how copyright law actually works.
And, from the emails, it certainly sounds like the publisher got permission from whoever has a much stronger claim to the copyright in most of the interviews it wanted to publish. And, of course, even if none of that is true, it seems like there's a fairly strong fair use case here, considering that it's an academic publication, and done as a compilation to look more closely at Kelly's interviews over the years (I wouldn't say that the fair use claim is a slam dunk, but there's a strong argument that can be made for fair use). But, of course, that requires Kelly to actually have a copyright interest (and to have registered it) in the first place.
But, really, let's take a step back here and look at the bigger picture. There is no legitimate copyright reason to grant Kelly a copyright in interviews that he did. He was not incentivized to do these interviews because of the copyright. He did them for whatever reason -- probably related to getting publicity in most cases. As such, it's ridiculous that we're even discussing a copyright interest at all here. There's no need for one for the interviewee.
So, in the end, this seems like yet another case of copyright as censorship. Patricia Ward Kelly does not want anyone else publishing a book that has extensive quotes from her husband (it should be noted that she's apparently working on her own book...), and thus the easy tool to use is copyright to censor this book that she doesn't control. Again, I can't see how this lawsuit survives very long, but it's another in a long list to add to examples of (1) copyright as censorship and (2) estates of deceased creators overclaiming copyright.
Compared to its early days, when releases of material like Collateral Murder dominated public discourse for weeks, Wikileaks is now only a shadow of its former self, eclipsed largely by Snowden's leaks. That's understandable, perhaps: Julian Assange has been holed up in the Ecuadorian Embassy in London for nearly three years, and it has been hard for the organization to raise funds to pay Wikileaks' running costs. However, that reduced visibility and activity doesn't mean it's not still releasing valuable material, particularly in the area of trade agreements. Today, it has published another interesting set of documents, this time from the field of surveillance:
WikiLeaks releases ten months of transcripts from the ongoing German Parliamentary inquiry into NSA activities in Germany. Despite many sessions being technically public, in practice public understanding has been compromised as transcripts have been withheld, recording devices banned and reporters intrusively watched by police.
WikiLeaks is releasing 1,380 pages of transcripts from the unclassified sessions, covering 34 witnesses – including 13 concealed witnesses from Germany's foreign intelligence agency, the Bundesnachrichtendienst (BND). The transcripts cover from the start of the inquiry in May 2014 through to February 2015.
WikiLeaks has also written summaries of each session in German and English as the inquiry, due to its subject matter, is of international significance.
This underlines an important aspect of Wikileaks' work: the fact that it seeks to make the documents it releases useful by providing commentary, summaries and indexes. Those valuable additions are often overlooked, but can play a crucial role in helping people understand the raw material provided.
The German parliamentary inquiry has been rumbling on for a year now, but has gained renewed importance with the recent revelations that the German spy service, the BND, has been searching through its databases using "selectors" (keywords) provided by the NSA, with apparently no oversight. Not only were many of the targets for those selectors EU citizens, but they included senior politicians and industry figures, too. Here's Wikileaks' summary:
One of the biggest scandals to emerge from the inquiry so far is the recent "selector" spy target list scandal where a BND official revealed that the agency was expected to spy on thousands of targets at the instruction of the NSA. These targets included members of the French government and European industry. This put into question Germany's suitability in taking a leadership role in the European Union. It also showed that international co-operation on mass surveillance, which has been marketed in public as a counter-terrorism measure, is in practice also used by the United States for the purposes of industrial espionage and geopolitical advantage vis-a-vis members of the European Union. The committee requested the full "selector" list of targets provided to the BND by the NSA. The committee was told that the US would first need to be asked permission for the list to be revealed to the committee (even in confidence). Last Wednesday, 6 May 2015, when the answer was meant to be delivered, stalling tactics were used, leaving the German public, and the Parliamentary inquiry, without any ability to understand what their own secret services are up to.
The "selector" scandal has now reached the highest political echelons in Germany, with Angela Merkel's earlier outrage over NSA spying -- not least against herself -- looking hypocritical at best, or dangerously naive at worst. Wikileaks' latest release therefore comes at just the right moment for those seeking to understand what has been going on in Germany. It's also a timely reminder that Wikileaks is still able to perform an important service in this respect, despite its straitened circumstances.
As was noted here earlier, the FBI took a bold step in towards joining the 21st century by finally implementing audio and video recording hardware introduced in the 20th century. Up until this point, the FBI, along with the DEA and ICE, did not record in-custody interrogations using anything more up-to-date than pen-and-paper. This rendered recollections of interrogations completely suspect, prone to pen-wielder bias and the insertion and removal of context as needed, presumably in order to help secure more convictions for the FBI's entrapment counterterrorism task force.
And, as was also noted, the DOJ's new instructions provided plenty of escape hatches for agents who wished their interrogations to remain as analog as possible. Unrecorded interrogations can still be performed in the event that desirable recording equipment (i.e., a cellphone) isn't available or if the equipment available isn't functioning (batteries missing/unplugged/inadvertently smashed to pieces…).
First, there's the "public safety" exception, which can be triggered when exigent circumstances make unrecorded and (un-Mirandized) interrogations a necessity. These would be questionings normally done in the first few moments of an arrest. But with everyone carrying around a recording device, that exception no longer makes much sense. You no longer have to take a suspect "downtown" in order to record a questioning. The inclusion of this loophole is likely borrowed from pre-existing language, but all it does is create reasons not to record.
[S]ince recording is no longer impracticable, why wouldn't a responsible law enforcement agency want to preserve an unambiguous record? Unlike a public safety exception to Miranda, a public safety exception to recording seems to serve no purpose other than that of affording a loophole that can be exploited for illicit purposes.
The other loophole is much, much larger. It's predicated on the same rationale that has allowed the Constitution to be selectively scrapped over the past dozen years.
The same point applies with even greater force to the exceptions for “national security” and “intelligence, sources, or methods.” If recording is feasible (and that is the only condition in which the recording policy applies), national security and counterterrorism officials can only gain by having an unambiguous record of precisely what a suspect was asked and precisely how he or she answered. Indeed, an official who deliberately chose not to make and preserve a clear record of a national security interrogation would display less dedication than incompetence.
As Schulhofer points out, this exception plays right into the mindset of the FBI, which has refashioned itself into the nation's largest counterterrorism force (putting law enforcement on the back burner). This also plays right into every law enforcement and intelligence agency's fetishization of "intelligence, sources or methods." This is what's conjured up to justify refusals of FOIA requests and to keep new surveillance methods out of the public eye for as long as possible. It's what's used to deny access to returned warrants on closed cases. But for the FBI, it's also a reason to never record anything, just in case. The FBI's intertwined relationship with the NSA -- combined with the last year of leaked documents -- will make any agent extremely wary about leaving behind undisputed records of intelligence-related interviews. But all this will do is make these agencies even more insular and untrustworthy than they already are.
No national security establishment can possibly operate effectively on the basis of unwritten knowledge and word of mouth. If our government has reacted to the Snowden affair by developing an aversion to writing anything down, we are in deep trouble.
"Deep trouble" is where we're headed, if we're not there already. The DOJ has given the FBI, DEA and ICE huge exceptions to the recording policy -- which, it must be noted, aren't actually commands but a "presumptions" -- ones that are particularly prone to exploitation. Over the past decade, we've seen the government exploit the fear of "the next 9/11" to expand power and contract civil liberties. Government agents may now have to act under the "presumption" that custodial interviews will be recorded, but the DOJ has given them a handy list of excuses to use when these recordings fail to happen.
There was no news release or news conference to announce the radical shift. But a DOJ memorandum obtained by The Arizona Republic spells out the changes that will begin July 11.
"This policy establishes a presumption that the Federal Bureau of Investigation (FBI), the Drug Enforcement Administration (DEA), the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) and the United States Marshals Service (USMS) will electronically record statements made by individuals in their custody," says the memo from James M. Cole, deputy attorney general, to all federal prosecutors and criminal chiefs.
"This policy also encourages agents and prosecutors to consider electronic recording in investigative or other circumstances where the presumption does not apply,'' such as in the questioning of witnesses.
As you can see, there's still a loophole for these agencies to exploit. They'll be "encouraged" to record non-custodial questioning, but there's no direct stipulation requiring it. On the plus side, this is a huge improvement over the previous method: handwritten notes taken by agents on a "302 form." Not only were the notes highly subjective, but they also tended to be destroyed as soon as possible. Others weren't even transcriptions of interviews, but were written after the fact using nothing more than an agent's power of recollection. This has obviously led to abuse, as well as to plain old carelessness.
The failure to maintain electronic records of interrogations also created gaps in FBI intelligence gathering, especially involving terrorism cases. Instead of maintaining an accurate and largely indisputable record, agents on the witness stand for decades have relied on their memories, interpretations and handwritten notes transcribed into a form known as the 302.
Critics have said that flawed system results in botched investigations, lost evidence, unprofessional conduct and false convictions. They noted that the historic DOJ practice was problematic in trials of suspects like terrorist Osama bin Laden, TV star Martha Stewart and Oklahoma City bombing defendant Terry Nichols, along with thousands of defendants with no public exposure.
Of course, those who prefer the old, unaccountable way are still raising objections, despite the fact that recording custodial interviews has been common practice in less-cutting edge local law enforcement agencies for decades.
Nancy Savage, executive director at the Society of Former Special Agents of the FBI, said there's probably no unilateral view from field agents. Although tape recorders sometimes intimidate suspects, she added, the change was probably inevitable because juries have come to expect audio and video evidence.
The FBI's other rationale for pen-and-paper-only has been that jurors might be persuaded to acquit someone if they were made aware of tactics used by agents to secure a confession.
In 2006, the New York Times uncovered another explanation for the DOJ policy, spelled out in an internal FBI memorandum. Basically, it argued that jurors might be offended, possibly to the point of acquitting defendants, if they observed the deceit and psychological trickery legally employed by agents to obtain information and confessions.
This would seem to be a problem the FBI needs to solve, rather than just expect the public to go along with its your-word-against-ours policy. The fact that the DOJ's several agencies need to be pushed into providing something more verifiable than an agent's slanted recollection of an interview is a sad statement on the mentality of those employed by these agencies. For them, it's been a long, easy ride, where any abuse under the old system could be mitigated by the agencies' "fighting the good fight" excuse. After all, they're chasing down drug dealers and terrorists. Who are we to question their methods?
Well, we're the public and we've been expecting accountability from our government agencies for a long time. And they've been in no hurry to provide it. By "boldly" pushing the ATF, DEA and FBI into the last half of the 20th century, Deputy AG James Cole is finally nudging his charges in the right direction. To be sure, recording devices can be tampered with, misplaced or simply never activated, but it's still a huge step forward from the agencies' long-held preference for pen, paper and subjective portrayals.
A bunch of folks have alerted me to this bizarre story of author Bill Bryson, known for his "humorous books on travel," along with his publisher Transworld (a division of Random House), who are apparently claiming that Bryson holds the copyright on an interview he gave 20 years ago to Mike Gerrard, who published it in the literary magazine Passport (where Gerrard was the publisher). Thinking people might be interested in that interview, Gerrard (smartly) dusted it off and republished it as a Kindle e-book (it sounds like a Kindle Single, but that's not made clear anywhere). Amazon has since taken the book down. No one seems to have published the actual takedown letter from Transworld, but the initial story on it does say it's a copyright claim:
According to Bryson's publishers Transworld, a division of Random House, you may not. Transworld wrote to Gerrard stating that Bryson did not authorise additional publication of the interview, or sale as an e-book beyond the contemporaneous interview for the magazine.
Furthermore, Transworld's lawyers argue that Bryson remains the owner of the words spoken by him.
That last line is not true. As plenty of people have pointed out, if that were true, it would basically make journalism impossible, because anyone who didn't like a quote could demand it be taken down. That's not how copyright law works.
Furthermore, Gerrard has been pointing out repeatedly that many of Bryson's own books involve him quoting others, and jokingly wonders what would happen if one of those people sought to claim copyright on one of Bryson's books. From the initial story in World Travel Market (linked above), it appears that Gerrard is willing to fight back over this claim. He's talking to lawyers to better understand what he can do.
Of course, Gerrard oddly quotes a US legal precedent against having a copyright in your own words, but it appears that both he and Bryson are based in the UK, so I'd imagine that any legal discussion would be under UK law which, tragically and stupidly, doesn't have fair use like American law has. Still, it does have fair dealing, though that should only come into play if Bryson has a legitimate copyright interest in his spoken words, and that seems unlikely -- though, having some UK copyright experts chime in would be nice. From a quick glance, it looks like there's no legitimate copyright interest here, and even if there was, there would either be a fair dealing defense for the purpose of journalism, or an implied license or some sort in the granting of the original interview.
Either way, it seems like yet another attempt to abuse copyright law for censorship, rather than any legitimate purpose.
FBI agents always interview in pairs. One agent asks the questions, while the other writes up what is called a “form 302 report” based on his notes. The 302 report, which the interviewee does not normally see, becomes the official record of the exchange; any interviewee who contests its accuracy risks prosecution for lying to a federal official, a felony. And here is the key problem that throws the accuracy of all such statements and reports into doubt: FBI agents almost never electronically record their interrogations; to do so would be against written policy.
Without a recording to compare the transcript to, we are expected to trust the FBI's version of the interrogation. If we can't trust it, we are left to draw one of the following conclusions.
1. The transcript is completely false. 2. The transcript is heavily editorialized. 3. The transcript interprets certain statements, but is otherwise accurate. 4. The transcript is completely accurate.
Of all of these choices, number 4 seem least likely. In fact, one wonders why the FBI bothers interviewing anyone when it could simply put two agents in a room and allow them to bang out a confession on behalf of the accused.
If a suspect claims the transcription is erroneous, it's his word against theirs. His words, of course, disappeared into the ether as soon as they were spoken. The FBI's version lives on, printed on paper.
We don't need to ask "why" this is a problem. There are rhetorical questions and then there are stupid questions, the sort helpful teachers and guidance counselors continue to pretend don't exist. A better question is, "Why hasn't this been changed?" Silverglate notes this policy is an updated version of a 1990's policy, crafted in 2006, long long long long after recording devices were ubiquitous. The excuse that this policy was "logistically necessary" because of technological limitations was ridiculous in 1990, much less 16 years later.
This is a problem. More specifically, this is Robel Phillipos' problem.
Phillipos is a 19-year-old Cambridge resident, former UMass Dartmouth student, and friend of alleged Marathon bomber Dzhokhar Tsarnaev. He faces charges of making materially false statements during a series of interviews with FBI agents. If convicted, he could get up to eight years in federal prison and a $250,000 fine.
Q: We found files on your computer showing that you went to a website with instructions on how to make a bomb, so we know you did it. When did you first go to the bomb website?
A: I surf the web constantly and go through, like, a million pages. I have no idea what pages I searched or when. How could I possibly know?
Notated in 302: D cannot recall when he first went to bomb website. Went "constantly."
Slick, isn't it? And when someone points out a misquote, the accusation is turned on them just as easily. "Are you lying now or were you lying earlier?"
This is nasty business but it gets even nastier. Beyond the hilarious claim that tech simply hasn't advanced enough since 1990 to allow reliable voice recording, there's a much darker rationale guiding this ridiculous (and dangerous) policy.
The more honest — and more terrifying — justification for non-recording given in the memo reads as follows: “. . . perfectly lawful and acceptable interviewing techniques do not always come across in recorded fashion to lay persons as proper means of obtaining information from defendants. Initial resistance may be interpreted as involuntariness and misleading a defendant as to the quality of the evidence against him may appear to be unfair deceit.” Translated from bureaucratese: When viewed in the light of day, recorded witness statements could appear to a reasonable jury of laypersons to have been coercively or misleadingly obtained.
Sometimes the "reasonable jury" would be right -- the statement has been "coercively or misleadingly obtained." Other times, it may not be as clear-cut. But in a day and age where recording interviews and interrogations is the expectation, the FBI continues to play by its own (convenient) rules. And if the person being interrogated doesn't like it, he can expect additional charges to brought. This puts the alleged criminal in the unenviable position of having "anything he says" twisted, rewritten and heavily paraphrased before being used against him.
Silverglate cautions to withhold judgement on Phillipos until all the facts are in. But as long as the FBI continues to use this "recording" technique, don't grant its statements any credibility. They have none.
In 2009, we wrote about the city of Bozeman, Montana asking job applicants for all of their passwords to social networking sites, so that the city could look more closely to see if they had "high moral character." Public outrage over that plan resulted in the city dropping the requirement. Last year, we noted that some places in Maryland were doing the same, leading to a lawsuit. However, the Boston Globe is now reporting that more and more jobseekers are being asked for their social network logins, as a simplified "background check" by employers. It's a newspaper trend piece, so in typically maddening fashion it makes no effort to indicate how widespread this really is -- but the fact that it's not a big story any time an employer does this certainly suggests that it's becoming at least somewhat more standard.
Still, does that mean we should pass a law? Senator Richard Blumenthal -- who has long been in favor of laws against all kinds of internet companies -- is apparently working on exactly that kind of legislation. Somehow, I doubt it's an accident that the Boston Globe trend piece came out at about the same time as Blumenthal's plans were discussed. Frankly, I still think that it's pretty sketchy and questionable for companies to ask for logins, but is it so bad that we need a law? Is there at least some sort of data on how widespread this practice is?
In the meantime, for those interviewing for jobs who do get asked for such things, it seems only proper to respond as the first individual profiled in the original article does—by walking away:
Bassett, a New York City statistician, had just finished answering a few character questions when the interviewer turned to her computer to search for his Facebook page. But she couldn't see his private profile. She turned back and asked him to hand over his login information.
Bassett refused and withdrew his application, saying he didn't want to work for a company that would seek such personal information.
I would still guess this isn't quite as common as the article tries to suggest, but either way I'm curious if people feel this practice is so egregious that it needs a new federal law?
While I have some very serious concerns about the way the UK's collection agency PRS for Music conducts its business when it comes to threatening small businesses -- including going after a woman playing music to her horses and woman singing while stocking the shelves at a store -- over the past few months I've been having a series of interesting conversations with Will Page, the Chief Economist for PRS.
Page, of course, put out that famous report last summer, that pointed out that the music industry in the UK appeared to be getting bigger, not smaller (contrary to what you hear from many people). Page is a fun guy to talk with about music industry economics, and we decided to run a little interview with him here. There's plenty that I disagree with him about, but plenty that we agree on too. There's so much in this interview that I'd like to dig deeper on, and I hope to do that in a series of posts in the future -- and some more back and forth with Page -- but I figured at this point it was worth getting our discussion as it stands out there for people to read.
We wrote about your study last
year showing that the UK music industry was actually increasing -- contrary
to most of the headlines were saying. Can you give a quick summary
of why your numbers show a very different story than the popular press
'different story' to what people are accustomed to simply because,
for too long, people have characterised the music industry as being
about just the recorded music industry. That's largely due to the
fact that the only data out there for people to discuss is recorded
music statistics. When we published 'Adding
Up The Music Industry for 2008' last year, it was an important step towards showing (i) how much the
whole music industry was worth and more importantly (ii) how it all
hangs together. One of the many audiences we aimed this work at was
Government, who need to understand the broader picture of what the music
industry comprises of, and the value that it brings.
The Insight paper allowed two new pieces of the pie to be illustrated
and properly understood: firstly live music revenues of
£1.4 billion and secondly business-to-business licensing revenues which
were over £900 million. From a total pie worth
£3.6 billion, that implied that recorded music made up a nudge over
a third of the total revenues -- that's a significant sum, but definitely
not the only show in town.
Both of us are
skeptical that the digital music sales market will ever replace the
physical music market. Can you summarize why and what numbers
you've seen about the digital market?
My concerns about the digital market
start with the same word that introduced me to economics:
'scarcity' -- there is little scarcity in selling digital media
goods and that inevitably affects price. I think the best way to illustrate
this is to look, instead, at the live music industry as those folks
are the masters of pricing scarcity
-- they view tickets as 'lots' and want to maximise the willingness
to pay for each. Live music mastered their demand curve a long time
ago; digital music is still trying to discover theirs. Another problem
with this topic is that interpreting digital music revenues is not a
straightforward exercise, especially in Europe. We published a paper
and Interpreting the Digital Market' two years ago to help folks try and get their head around this
complex market, and it's not got any easier since!
I'd like to flag two observations
for your readers. Firstly, don't view digital in isolation, when it's
shown that one-in-five albums sold in America were digital, that tells
you more about the collapse in the five, that the outperformance of
the one. Secondly, the UK has really outperformed its European neighbours
in developing a large, and importantly diverse, digital market. UK digital
revenues per capita are twice, maybe three times, that of our main European
neighbours, which is a great testimony to the work that Jez Bell at
PRS for Music and folks like Francis Keeling at Universal have done
on the licensing front as well as the
incredible achievements of the services
like We7, Spotify and 7 Digital which have taken out the licences and
Finally, whilst the digital makes
up 20% of recorded music revenues, and 5% of PRS for Music collections,
what I really have learnt to appreciate is that these digital services
are legal 'venues' -- a concept that Eric Garland drilled home
to me -- and somewhere north of four million
folks in the UK are going to sites like
Spotify or We7 and doing their thing
-- now it might not be producing the monies people once wished for
but they are arguably not going to Mininova, an illegal venue, and that's
an important achievement -- especially when Mininova celebrated its
10 billionth torrent download
three months before iTunes celebrated theirs. Engagement with legal
venues is worth more than the top line
revenues might initially suggest.
You mention 'scarcity' in the context of live vs. digital, but live has a real scarcity (seats -- over which they can control access). Digital doesn't have that kind of scarcity. You say that digital hasn't 'discovered their demand curve,' but might the bigger issue really be that without the scarcity the supply curve is the issue? My view has always been that the digital market is a red herring due to the lack of scarcity, but instead the music world should focus on external scarcities that widespread digital music creates (including things like seats at concerts). Is the real issue not the demand curve but the supply and the failure (of some) to recognize that they need to think broader in terms of what they're selling?
That's a very insightful question -- and you're right as one of the many mistakes economists make is to forget about the supply side dynamics of a problem, and instead focus on demand. It's worth citing Jean Baptiste Say, and his Law of Markets which is that "supply creates demand." What this means, with regards to your question, is that "overproduction" in a free economy is actually impossible. That's a controversial proposition though, as I think it comes up against another trade off which we could call the attention economy, where a wealth of information leads to a poverty of attention. Stepping back from the theory, there is clearly more noise in the market place -- more artists, more songs, more places to hear them -- therefore more investment is needed to stand above the noise, to enable the benefits of your 'external scarcities' to kick in. One final piece of twisted economics is this idea of a 'freemium' model, which is cool but has a flaw -- if everyone did it, the less successful it would be. Point being there would be more noise in the free market, which erodes the value of the premium offering -- an increase is supply depresses price, and we shouldn't lose sight of that basic principle.
You call services like Spotify and We7 "legal venues" and things like Mininova as an "illegal venue," which I assume many of our readers may have an issue with -- especially given that Mininova has long had a program for artists to offer up their own content, and there certainly are a small, but growing, number of artists who have embraced those venues for legitimate marketing reasons. Is there an argument to be made that, given the size of some of the userbases of those venues that you (and many) deem illegal, that there may be ways to embrace and engage with them, rather than write them off as such?
The best way to embrace those users is to ensure the services they use are licensed and respect the value of music. Now, we have over a thousand digital music licencees here in the UK, and we've been granting online licenses since 2002, long before iTunes -- a fact often overlooked. The best way to approach the unlicensed services is to think of it this way -- we're all chatting about whether Spotify will sink or swim, right? That's the hot debate at the moment. Well, I would argue that at the margin Spotify would have far more chance of swimming, or up selling the subscription service, had they not had to face this unfair competition of illegal free. That's a powerful argument when you run it through, as it moves away from the old arguments and towards a more plausible observation: what opportunities are being foregone in the legal digital market due to the unfair competition of illegal free? One last thing on Spotify, which is that they went legal before going popular, bucking a regrettable trend. When you explain that to an emerging artist or songwriter, offering a counterfactual of many other sites which have become incredibly popular (and then flipped for incredible amounts of money) before taking out licences -- it really hits the message home.
Both of us are still
quite optimistic that there's still a huge opportunity for the overall
music market to grow. Where do you think that opportunity exists
-- and why is it mostly ignored?
If we pick up on
'Business-to-Business' revenues, or licensing income, this makes
music free at the point of consumption with compensation taking place
elsewhere. This part of the music industry is likely to make up an increasing
part of an increasing pie, and that by default presents opportunities
However, what's frustrating is that music licensing is an area of
the industry that's often least understood by emerging bands
and songwriters -- the MySpace generation --
who are trying to get one foot on the ladder to success. To realise
those opportunities, the first thing artists and songwriters
need to do is protect their rights by joining PPL and PRS for Music
in the UK, or their equivalents in their respective territories. Secondly,
it's very important that the licensing bodies
around the world get involved with the artists and songwriters. Here
at PRS for Music, we've got Myles Keller leading our membership development
and we're getting increasingly involved with our songwriters through
programming events like The
Great Escape on the 13th May, whereas in America you have
'walking encyclopaedias' like Todd and Jeff Brabec who are very
accessible on the panel circuit and their bible 'Music Money and Success' is required reading. I guess my point is
that the best way to realise the licensing opportunities that exist
is to get involved. Passivity doesn't pay.
You stress the importance of getting people to sign up for collection societies/licensing organizations -- which isn't surprising, given your employer -- but myself and many of our readers are concerned about the incentive structures when musicians rely on such organizations (even when-- as in many cases -- they're non-profits). With such organizations, you can take away some aspect of market-pricing, especially when there are issues of compulsory licensing and/or only one provider in the market. It also creates situations where those organizations constantly push for greater rights, or the ability to collect from more places for more reasons-- often upsetting other aspects of the market (for example: bars and restaurants no longer letting bands play live or hosting open mic nights to avoid having to pay licensing fees). While I agree that, given today's setup, it makes sense for musicians not to pass up revenue that's there for them via these organizations, isn't there a risk that these types of organizations distort the market from a purely economic viewpoint?
Each collecting society is different, so firstly -- let's be wary of generalisations. In America, for example, you are absolutely unique in that you have competition within collecting societies with ASCAP, SESAC and BMI -- the latter which is owned by the broadcasters! Similarly, the story behind SoundExchange is unique too -- and in many cases the US is playing 'catch up' with the rest of the world when it comes to neighbouring rights. So, I just want to be clear for a predominantly US Techdirt audience, the US experience with collective rights organisations will be unlike anywhere else in the world. I really mean that too -- it's such an exception to the rule.
Now to your question -- let's start by asking what is the rational for collecting societies. I would argue that the answer is three-fold: (i) reducing transaction costs for both rights holder and user, (ii) preventing fragmentation and (iii) solving co-ordination in many-to-many markets. The bottom line is this: PRS for Music enables start ups to start up, and songwriters to get paid. If you wiped the board clean and tried to devise a new model, which can hold together a blanket licence and balance the needs of unprecedented digital services, you would probably end up with what PRS for Music is doing just now. It's not an easy task, and armchair critics would do well to consider the complexity in this two-sided market and the trade-offs that we face every day, but to read that We7 now feel that add funded music can add up is heartening as it suggests we're getting this delicate balancing act right.
You've noted that the UK music
market appears to have again gone up in
2009 over 2008 and appears to be growing faster than other countries.
Why do you think the UK market has been different than elsewhere?
Firstly, The UK is not alone in bucking
the downward trend as Sweden, Denmark and Australia can also claim to
be outliers in some form. However, these are the exceptions as
opposed to the rule, and it's a stark contrast to the downbeat sentiments
I'm hearing from the US, and chalk-and-cheese to the situation in
Spain which really is frightening on many levels. I'd offer three
exceptions which have bucked global trends rule. Firstly, the live music
industry has continued to exhibit robust growth in the UK even in the
middle of a credit crunch, whereas other territories suggest the market
might have matured. Secondly, UK labels have arguably done a better
job of diversifying their revenue streams , due in part to the success
story that is PPL, and I doubt that level of diversification is being
reflected by labels in many other regions. Third, the UK really values
music. It's a simple point, but it really matters. Think: the role
of the BBC in championing emerging bands, the explosion in music festivals
in every corner of the country, the insane amount of work of Feargal
Sharkey at UK Music has put in to get all the stakeholders (including
ISPs) to banging heads together to face up to the challenges -- all
these ingredients help illustrate that this thing called "music"
actually matters to the UK. Conversely, I'm spending an increasing
amount of time in Spain now, and what you see there is that music doesn't
matter as much...if at all. It's one of the few western countries
that can claim a thriving digital AND physical piracy problem and investment
in domestic talent is drying up as there simply no return. It's actually
kind of eerie when you compare the quality of debate and level of activity
being had in the UK to that of other countries, it's not that we've
solved all the problems, far from it, but it's more about not dodging
them and actually doing something about them.
On Spain, I know the IFPI's recent report said the industry is in trouble there, and you do the same here, but we keep hearing from people who claim otherwise -- that there's a renaissance of music in Spain due to more widespread ability to promote and distribute musicians. Anecdotally, in the last year, I've actually picked up (yes, legally bought, on CD) albums from a few Spanish bands. Do you have some numbers for Spain -- since between the two of us, we seem to have very different anecdotal experiences? Could it just be that the business models haven't adapted yet?
Neither of us is from Spain, nor do we currently live there -- so we have to work this one out based on our own anecdotal experiences. What I've noticed is that trade revenues of record labels have halved in less than a decade, in nominal terms. I've also noticed incredible resentment to 'paying for music' in Spain, there's a real 'stick it to the man' attitude which is puzzling. I come back to the point on domestic investment -- given the situation there, would you (and that could be a label, publisher, manager or third party) invest in developing domestic talent in Spain, or would you invest somewhere else with a lower risk profile and then import into Spain. I'm sure there are lots of opportunities down there on the ground, but how many of those opportunities lead to a sustainable living for professional artists and songwriters. For me, Spain's situation is like a tipping point which other countries should take note of.
One point that you've noted is
that the live music market has grown and actually
surpassed revenue from recorded in the UK. Critics dismiss this,
claiming that the live numbers are dominated by "heritage"
or "legacy" acts. Is this true? You claim that
the revenue for live covers "more bands, more tickets, more seats,
more events." Who's right?
You're right to pick up on the changing
of the guard observation from last year, and it's incredible to think
that five years prior, live was less than half the size of recorded
-- which makes you ask three questions: (i) how has live captured so
much value, (ii) how has recorded lost so much value and (iii) is there
a link. However, read beneath the top line and you can consider the
distribution of those revenues: who got what share of the spoils.
As with recorded music, in live we're witnessing a hit heavy skinny
tail distribution, and that intuitively makes sense. The bigger you
are, the more forms of revenues you're able to exploit and the distribution
skews to the head naturally. When Take That played to over a million
people, that's an example of more tickets, more seats and more events
but it's just one band. Down in the tail, the picture is less clear
-- with worrying stories of support bands having to pay to play needing
to be balanced with the fact that the explosion in festivals gives more
opportunities for acts to get wider exposure. There's some interesting
signals coming out of the market place though, for example I was told
that there was noticeably less record label A&R presence at SXSW
this year, with agents and promoters filling up the bars on Sixth Street
-- perhaps that's a sign of the times.
You've noted in the past that
60% of the UK population don't buy music anyway and that "you can't
cannibalize zero" when it comes to things like file sharing "taking
away" revenue. Do you believe there's evidence that the 60%
of people who don't buy music are helping to contribute revenue elsewhere
-- and if so how and where?
It's a vital observation that needs
to be rammed home as the rights holders are understandably obsessed
with cannibalisation, but sometimes blinkered to the wider problem.
The legendary Rory
Sutherland remarked on
that "Can't Cannibalize Zero" phrase as a masterpiece and
told me that it reminded him:
"Of working with
ATOC, and First Great Western. They were obsessed with the risk of Revenue
Abstraction -- the rail phrase for cannibalisation. In other words, any
special off-peak offer was viewed with terror, lest it attract people
prepared to pay full fare. But, just like music, 60% of people don't
use trains - ever!"
Rory makes you think about the problems
differently -- and here, the problem is how can we re-engage the lost
majority? I collaborated with Spotify on a piece of research called
to dance to ARPU" which
allows rights holders and users to approach that infamous acronym with
more clarity. At the back of my mind, though, is this: most of the folk
of my parents generation love Spotify and none of them ever bought music
...ever. Engagement is the horse, and monetisation is the cart -- if
services like Spotify are helping re-engaging those who gave us nothing,
there's a better chance of getting something going forward.
Following up on the B2B side
of the market, some also point out that this part of the market may
also be dominated by large legacy acts who can score big sponsorship
deals. Do you think that's true and if not why not? What
opportunities are there for less well known bands in this area?
To quote from the paper, 'brands
investing in music trough sponsorship are drawn to it through the potential
audience affinity and reach; this means that much of the major expenditure
is biased towards the larger priorities and artists, which provide larger
fan bases.' That means that it's tough in the tail for bands wanting
to strike sponsorship deals. That said, there is a lot of scope to use
initiative to innovate in this sector. Here in the UK, we have organisations
like Music Ally and FRUKT who are doing some great work in this sector,
especially in terms of offering training and workshops for artists and
managers -- their material is well worth tracking as opportunities
in this sector don't find you, you have to find them.
Lots of people have suggested
that even if live is now outpacing album sales, it was still the record
labels that really financed tours and the growth of live. Are
there mechanisms to support and nurture live if the record labels continue
to decline? Where might it come from?
The kicker is this -- the money is
live is centred around the head, and much of that head is heritage in
status -- so the question I always ask is who's going to offer the
tour support for new bands to build the sort of fan bases that provide
the live industry with the heritage acts of the future. That's a legitimate
question to ask, and not an easy one to answer, but you've got to
look forwards not backwards, and I'm really hype on the company Songkick it's basically Facebook for folk who
love going to gigs with full functionality for ticketing, recommendations
et al. I think that what Ian Hogarth has done there is a real game changer
when you fully think it through -- and it also helps level a heavily
tilted playing field as emerging acts can benefit as much as the established
bands from Songkick's functionality. You have to manage expectations
as it won't make touring across a country in a bus sitting next to
a drummer with an odour problem any less unpleasant, but it does have
the potential to lead to more bands performing to more fans, and importantly
more data to build upon that success.
Notably absent from your discussions
on these numbers is anything (outside of live) having to do with direct-to-fan
opportunities that we've discussed on Techdirt. These numbers
may get mixed in elsewhere as they sometimes include
album sales and sometimes include live, but do you have any thoughts
on that market? Do you have any numbers on how those efforts are
Firstly, I've recommended your excellent Trent
Reznor case study to literally
everyone and their dog. What's really good about that is that you
echo what I've stressed every time I've explained 'In Rainbows, On Torrents' case study which was that this was a solution
for Radiohead, and was NOT a solution for the music industry. That said,
what Radiohead and NIN did were 'experiments' and we've got to
learn from these experiments. You got to ask the right questions --
so 'of what worked, what's transferable?' Second, whilst Topspin
was behind your Trent Reznor case study, there's another Toronto based
company worth checking out called Official
Community. They're providing
direct artist to fan infrastructure which allows for disintermediation
of the value chain, more empowerment of the artist and faster cash flow.
When you look into these models though, it's important to keep a balanced
perspective and manage expectations -- it's not going to change the
world, and it might not even change the actors involved in a 'conventional
deal,' but this existence of more options should, if anything, allow
artists and songwriters to negotiate better terms. Third and
finally, I agree with the premise of the question -- what's happening
outside the conventional radar is probably bucking the southward trend,
but because it's not being picked up -- the trend continues southward.
I learnt recently that the annual Cambridge Folk Festival is a massive
player in selling CDs of those folk artists to fans -- literally tens
of thousands of CDs being shifted on location. Now, you may be tempted
to dismiss this as just a niche festival and just niche CDs, but they're
shifting lots of them and there's are lots more similar festivals
up and down the country who are increasingly doing the same thing and
its questionable how much of this is getting picked up on the conventional
radar. That offers optimism for the future as regards the "known unknowns"
which are out there, but also presents numerous headaches for myself
and Chris Carey as we try to calculate this year's 'Adding up the
Music Industry' report together which is due for publication in July
You mention the "skinny tail" and that some of the success today is from heritage acts, but then we see numbers from folks like TuneCore that show a massively successful long tail. It makes me wonder if -- as TuneCore notes -- the long tail success stories simply aren't being seen in the data because it's the wrong data. We hear so many stories of musicians successfully embracing new business models down the tail that it makes me wonder if what's happening down there is simply not being counted. Thoughts? Any ideas on better ways to measure?
Both Chris Anderson's work on Long Tail, and our analysis since, has suffered a lot of misinterpretation because you can't dive into this topic and expect a simple tabloid headline to explain it all. Statistical distributions of large data sets are not the sexiest topic for the music industry to discuss -- on that we can both agree! But let's roll back to what I've stated repeatedly in our work here -- I loved the concept of the Long Tail, still recommend the book to colleagues and wish it would work the way we all hoped. However, it is a book about the supply curve -- here's what happens when lots of goods can get to market. What I was able to do, thanks largely to the mathematical guidance of Andrew Bud, is derive the demand curve for digital music -- which is like saying "okay, once you're on the digital shelf, who actually wants you." You need two curves to tango in economics, and we've been able to develop an unprecedented understanding of this digital music market place as a result. What's great though, is to know where economics needs to hand over to other disciplines, such as psychology, sociology or anthropology -- basically how do we understand culture.
I can illustrate what I mean by offering your readers a genuine exclusive -- by exhibiting the Lorenz curves for We7 and Spotify side by side, and comparing those with the sort of distribution Chris Anderson's theory predicted:
The red line is to show what a "great example of the Long Tail at work" should look like, where 95% of the niche inventory (reading from bottom left to bottom right) makes up 75% of the streams -- a fat tail. Clearly, neither We7 nor Spotify look like that, with both curves tugging into the bottom left hand axis point and this is what's meant by a hit heavy, skinny tail distribution. However, the curves are different, and that is to be expected -- as We7 has a strong editorial with excellent artist promotional campaigns, whereas Spotify is editorial free and allows the consumer to graze the field at their leisure. Consequently, you can see that We7 (blue line) is more hit centric with a 90/5 rule and Spotify (green line) is more democratic with an 80/5 rule which, when you step back, is common sense made complicated but it's nice to see the math adds up!
The key thing for Techdirt readers is that's what economics can tell us when rights holders and users collaborate to understand unprecedented markets, and it's great that PRS for Music and Digital Music Services are willing to work together like this -- I think it's a important part of the success story in the UK. However, economic analysis can only tell us so much and it's at this point when the baton must be passed on to folks from other disciplines or backgrounds who can bring new insights to the table to work out what that actually means in terms of this intriguing thing called 'culture' -- which also means this is a good point to conclude this interview.
Thanks to Will for this fun discussion... which I fully expect to continue. If you want to see one of Will's recent presentations on the state of the music market, it's embedded below:
While Metallica is claiming (this time around) they don't mind that the latest album is being shared online, someone forgot to tell the band's representatives. TorrentFreak notes that the band's European record label, Universal Music (in the US, the label is Warner Brothers Records) has canceled the band's scheduled interview with a Swedish newspaper after a reviewer for the paper reviewed a remixed version of the album that he admitted he had downloaded via The Pirate Bay.
Universal's claim was that this is "totally unacceptable" seems destined to backfire. This is only going to provide a lot more attention to the fact that there's a different remix of the album available online that many people find better than the official version. Plenty of other artists like and encourage others to remix their music, because it tends to get fans much more involved with the music and can help attract new fans while making existing fans more connected to the band. For all of Metallica's talk of really embracing online fandom this time around, it keeps showing that it puts up roadblocks all around, rather than really connecting with the fans.