Over the past decade, the music industry was roundly criticized for how it responded to the proliferation of digital content — launching hundreds of lawsuits against otherwise law-abiding consumers who downloaded music. Although many companies have backed off from that approach, the challenge remains to find a business model for digital music. If songs can be downloaded and traded for free, how do the musicians, recording engineers and record company executives get paid?
Jim Griffin, working as an advisor to Warner Music Group, has a plan to find an answer.
At the recent Supernova conference in San Francisco, Griffin — who spoke on a panel titled, “Is There a Media Business?” — characterized the music industry’s conundrum as “Tarzan economics”: “We cling to this vine that keeps us off the jungle floor. Yet, at the same time, we swing for the next vine, because we want to be propelled forward. The trick is figuring out when to let go of that old vine and when to embrace that new vine.” Griffin points out that the problem is particularly pronounced for those for whom the “old vine” is still profitable and who have a “responsibility to continue to book that revenue.”
Griffin, however, is looking for the next “vine” by working with the music industry to explore new revenue models for the digital world. Starting this year, he is partnering with a number of universities to experiment with business models that allow students to download and share music, while finding a way to remunerate the artists who create that music.
The initiative, dubbed “Choruss,” was started by Warner Music Group and seeks to become an industry-wide consortium of artists and recording companies whose goal is to develop a system of “collective monetization,” a means of generating revenues in a way that acknowledges the current state of affairs where music is easily traded online.
Knowledge at Wharton spoke with Griffin following his panel presentation to learn more about Choruss and how it might change the way music is distributed and monetized. An edited version of that conversation follows.
Knowledge at Wharton: What is your role at Warner Music Group, and how is it related to Choruss?
Jim Griffin: Warner owns Choruss; it incubated Choruss. Choruss is now becoming an independent company with Warner still involved. [Warner Music Group CEO] Edgar Bronfman saw the wisdom of a shift from product to service, and [thought]: “Yes, I will incubate that here at the company. You can start with our content and then seek to bring others in.”
We are looking at a transition from product to service, where that service requires a kind of aggregation that has occurred in the past in the music business. In some ways it predates the music business itself. It started in 1847 in Paris — at the Ambassadors [Les Ambassadeurs] restaurant on the Champs-Élysées. A famous composer, [Earnest] Bourget, went to dinner and heard his music being played by the orchestra. He left without paying for his meal because he said, “They’re not paying me for the music I wrote, so why should I pay them for the food that they gave me?”
He goes to court and the judge says, “You have to pay for the meal, but they have to pay you for the music, too. But good luck getting paid. What do you want? Do you want every restaurant, every place that needs your music to contact you? How are you going to get paid in this new world?”
At that same time and place, Victor Hugo and Henri de Balzac had a law passed that said that you couldn’t read their books aloud in public without paying them. They had this mutual interest: How do we get paid in a world where our works are distributed widely — [in a] restaurant or a hotel lobby or whatever? They formed the first collective for musicians and writers, what became known as SACEM [Société des auteurs, compositeurs et éditeurs de musique].
The concept spread out of France to Canada to the French-speaking territories and then to the United States in the form of ASCAP [American Society of Composers, Authors and Publishers], BMI [Broadcast Music, Inc.], and SESAC [originally, Society of European Stage Authors and Composers, although its current focus is broader]. It is a notion that we call “collective monetization” or “collective licensing.” And it has played a key role in the monetization of music over hundreds of years.
Knowledge at Wharton: Given all this history, why are we in such dire straits now? Why haven’t we worked this out?
Griffin: Humorous, isn’t it? It’s because what intervened was a product business in music — the idea that you could put music on a piece of plastic or a piece of vinyl. The service promoted the product, and everybody was happy.
Now we are beginning to realize that music — the product — exists only for a blip in time. It’s not [now] the dominant way that we get music. And yet it was how we monetized music for quite a time.
Now we get back to the notion that music is a service, not necessarily a product. And its financial value can be measured as the quantity and the quality of the crowd it draws. That leads us back to the idea of: How then do we collectively monetize music?
Knowledge at Wharton: How is Choruss going to provide a solution to this?
Griffin: Choruss is that very notion of a collective monetization society — much like that first one, SASEM. Music’s mindshare has never been higher. It’s not a buggy whip situation — it’s not like we don’t need it anymore. Music is present in almost every place we go. It still has an enormous value in keeping people in a restaurant ordering more drinks and getting more food, keeping them aggregated in hotel lobbies, drawing them to cable channels and websites.
[Music] is more relevant than it has ever been. Its mindshare has never been higher. But its market share has never been lower. That is a result of a transition I call Tarzan economics, a clinging to this vine of product but knowing that service really is the value. How do we fully realize the service value where there is, by its very nature, an aggregation occurring?
Knowledge at Wharton: You mentioned BMI and ASCAP, which are organizations that handle royalties for the publishing rights of music for songwriters.
Griffin: Yes. ASCAP, BMI, and SESAC only collect for the performance of music, not its recording.
Knowledge at Wharton: Should we think about Choruss as an attempt to provide a similar model for recorded music?
Griffin: Very much so. You should see it as sound recording companies learning to act more like music publishers — realizing that the financial value may not be fully represented in a manufacturing operation that controls the quantity and destiny of music, but instead in the economic value that music brings to any activity.
It is an interesting conundrum that today, if a sound recording plays on the radio, in a hotel lobby or in a restaurant in the United States, the songwriter gets compensated, but the sound recording owner does not. Yet without that sound recording there is nothing to pay the songwriter for.
Knowledge at Wharton: The examples you have chosen — of music played in a hotel lobby or a restaurant — are, in a sense, easy ones. The greater concern is about music downloaded over Limewire or BitTorrent. Are you hoping to cover that as well through this model?
Griffin: Yes, that is precisely our goal: To evolve, over a period of time, a model that works for performance and extend that to what were traditionally products on the ‘Net.
Knowledge at Wharton: What might this look like? Can you make this a little more concrete?
Griffin: Sure. In fact, we are working on those very experiments right now. There are a half a dozen schools that are conducting Choruss-related experiments that seek to uncover the degree to which this model can be extended to the downloading of music.
They are experimenting with different approaches. But let’s take the prime example — peered sharing of music — because we know that music fans love to share music. Here we take peer-to-peer sharing of music and we put a flat fee price with unlimited access. We take those flat fees and we put them into a pool and then divide those pools based upon metrics that are derived from the use on the network. [The metrics] might be: how often a song is acquired or how often a song is played. No one is sure. This is an art, not a science.
Knowledge at Wharton: How would you know how often the song is played?
Griffin: You could [survey] a representative group. Or you could ask them if they would agree to be monitored for a time — à la Nielsen or Arbitron — the same way we do it for radio or television. While it’s imperfect, it is a metric that works. It is good enough for commerce. We have a very robust advertising economy that moves on what people say they watch or listen to. It’s good enough. We bring that same approach to the downloading and the sharing of music in some of our experiments.
Knowledge at Wharton: Do the experiments that are underway at the universities include only Warner content?
Griffin: No. We’ve asked all artists to participate. The other record companies are participating — and publishers and songwriters as well.
The idea is to test the notion that you can pay for music as a service and that money can be pooled and divvied up in much the same that we do with cable [TV], for example.
Knowledge at Wharton: One can see how this might work in a university where, like your cable TV example, there is some control over the distribution network. But in the case of the global peer-to-peer networks, you don’t control the distribution point.
Griffin: But you needn’t. There is no control necessary under this model. All one needs is the payment of a fee and some metric for its delivery to continue to incent the creation of content.
Knowledge at Wharton: How does the payment of a fee happen? Who pays?
Griffin: In this case it is a voluntary decision that an individual makes: Do I wish to participate in this system? They are then given an account and a password to trade all of the music they want, knowing that the activity is legitimate, that they have paid a fee for doing so, and that the fee will be allocated to those who own the rights for that content.
Knowledge at Wharton: This would be opt-in?
Griffin: Sure. Like cable. You get to decide: Do you want to subscribe to HBO? And when you do, the same thing that happens — your money goes into a pool that is used to pay for rights.
Knowledge at Wharton: Yet, in the case of music downloading, there is an embedded culture of free music sharing.
Griffin: Sure — because there is no legitimate alternative. In other words, the point is: Can we get at the motive for piracy without worrying about its mechanism?
Traditionally the way of dealing with piracy has been to attack the mechanism. Here, there is an acknowledgement that technology not only has made it voluntary to pay for music, but that every year it is going to be more voluntary. If we reach that understanding — that, as a technical matter, it is voluntary to pay — then our question becomes: How can we deal with the motive for piracy?
There I think the response is a business response. It is not the law. It is not technology. Traditionally our approaches have been law and technology and they have proven to be clumsy and blunt tools indeed.
The point is: Can we make a business deal with an end user that makes it more attractive to pay? Can we give them a better service than piracy can give them? Can it be faster? Can the files be better quality? Is there a price that you are willing to pay for that? We are experimenting with that very question.
Knowledge at Wharton: Is a payment scheme part of the experiments you’re now launching?
Griffin: Yes, absolutely. They have to pay.
Knowledge at Wharton: Who pays: the university or the individual students?
Griffin: The individual has to pay. You wouldn’t be learning much in an experiment if someone else was paying for your music. Everybody would be for it, right? That wouldn’t be a good test.
Knowledge at Wharton: You have not identified the universities that are participating — is that right?
Griffin: Right. For right now they have asked for their academic privacy. We don’t ask them to stay private — it is up to them. [Because] students are involved and it is an experiment that they are running, [the schools] ask us not to give out their names. But if they want to, that’s fine…. They are designing their programs now. It takes time.
What we have done is empowered them to search for a solution around collective licensing because this is something that they can only do when the licensing is being handled by someone else, because it is very difficult for them to get together and to license music.
Knowledge at Wharton: At this stage, you don’t know things like what percentage of students are opting in?
Griffin: Don’t know yet. [The programs haven’t] started [as of the date of this interview]. I did say publicly — and I stand by it — that tens of thousands of students have indicated their willingness to pay because their school has come to us and said, “Look. We are willing to pay on behalf of all the students.” And we have said, “Well, you can’t pay on their behalf, but if you can collect money from them individually — that is fine.” Those schools have come to us and said, “We are willing to pay for everyone on our campus to get all the music they want.” It’s important to be point out in this debate is that students are willing to pay and groups of students are coming forward to say, “We are willing to pay en masse.”
Knowledge at Wharton: Since you brought up your critics — you do have some.
Griffin: Oh, absolutely. I value that. I think it’s wonderful. There are lots of critics. But they are not critics of what we are doing because they don’t know what we are doing. It enrages them that they don’t know.
Knowledge at Wharton: In fact, that is one of the criticisms: that you’re working in secret and you then evade specifics by saying, “It’s just an experiment.”
Griffin: But we’re dealing with schools. We didn’t come in with this attitude. The schools took us aside and said, “If want to work with schools, you have to respect academic privacy.”
Knowledge at Wharton: Once the experiment is underway can you — will you — talk about the aggregate outcomes and what revenue models seem to work?
Griffin: Absolutely. It is up to those schools, because it is their results.
Knowledge at Wharton: Some of your critics refer to what you’re proposing as a “tax” on music downloads
Griffin: That is ridiculous. They say it because when we first started talking about Choruss, Wired magazine called it a tax. The assumption was that we would require everyone to pay. Whenever you discuss compulsory licensing or collective licensing, some people think that it is going to be required.
There is a huge difference between requiring artists to be involved and requiring people to pay. And we are not dealing with compulsory licenses — it is voluntary both on the part of the rights holder and on the part of the student.
But when you simply enter into a discussion around compulsory licensing there is the belief that it is somehow going to be mandatory for the rights holder, or the musician, or the individual to pay. In our case that’s not true. We don’t require any student to pay. And we don’t require any artist to be in.
Some countries do use compulsory licensing. And there are some compulsory licenses in the United States — for example, for webcasters. Every artist has to permit their music to be used by a webcaster. They don’t get to decide. The government has adjudged that if you have released music commercially it can be used by a webcaster. That is a compulsory license. By no means does it require every webcaster to pay in that fashion. They could reach their own licensing agreements. And neither does it require individuals to pay webcasters. But it is compulsory upon the artists to permit their music to be used in that fashion.
And so it is easy to see how the debate gets lost and then someone says, “Oh, if it is a compulsory license then I am going to be compelled to pay.” That certainly doesn’t follow, but I see the misunderstanding.
What could be a tax about Choruss when the individual gets to decide if they pay and the artist gets to decide if they are in? That can’t be a tax, right? But in a headline it makes people read an article. If you say “music tax,” you will get a lot more hits on your article than if you say they are thinking deeply about how to get music paid for.
Knowledge at Wharton: If all goes well with your experiments, describe the situation in five or ten years in the future.
Griffin: They have already turned out well. This is not just a test of how the market perceives music. It’s also a test of how rights holders come together to make that possible. And that behind the scenes work has been going on for two years, and we are learning a great deal about it.
Let me give you an example of the things we have learned. When we started this experiment we were thinking about premises licensing. ASCAP, BMI and SESAC license premises or a radio station. They say to a restaurant or a radio station, “You can use our collection of songs.” Our thinking was, okay, we could go to [an institution’s computer] network — like we go to a pub or a radio station — and say, “On your network you can use the following collection of music in any way you like for your users.”
We started talking to students and they said, “I don’t want my music service to stop at my network’s borders.” We realized that 80% of the students live off campus. They have jobs off campus. They go home over the holidays. They don’t want their music service to stop at their network’s borders. Their music service needs to traverse network borders.
We immediately changed our thinking. We learned right away that monetizing networks isn’t the point. Networks may be the locus of the activity, but it is certainly not the solution in terms of monetizing the activity.
That is an example of learning that immediately gripped [us] to say: Okay, it’s not about monetizing a network. It is about creating a service that is monetized. That service can traverse networks.
Another example was: Some rights holders were very clear with us that they felt that in the future music money like this should be allocated on plays. They didn’t come to us with a degree of moral certitude, but they said, “We would like for you to experiment around new metrics for the allocation of the money that is collected.”
In a retail world, let’s say, you and I buy two records. One of those records we play a lot. And the other record we play not at all — we didn’t like it. Those artists got paid the same. Maybe in a future digital world the artist who gets played more should get paid more. Maybe. That is part of our learning. How does one figure that out? How do you know which music is played most often?
Knowledge at Wharton: How do you know how often a song gets played in a way that’s not intrusive to the user?
Griffin: I think most people don’t feel intruded upon by, say, Arbitron or Nielsen. Some people do. But I think when people get that diary in the mail along with the buck they honestly want to figure it out. While those ratings are flawed — and inherently so — they are good enough for commerce. We have an industry that rewards some performers better, some disc jockeys better, some announcers better, some networks better — because they have a better quality or quantity of crowd. So we do use those imperfect measures a lot.
Knowledge at Wharton: So some of your models are looking at differential payment of artists?
Knowledge at Wharton: But you are looking at a flat rate payment from the consumer — whether downloading one song or a thousand in a month?
Griffin: We are experimenting [both ways]. Because it may be that in some cases you would prefer to buy a niche as opposed to everything. You might say, “Jim, I’m a jazz fan. I don’t want to pay for polka music.” It may be that it is a more attractive offering to you to have a flat fee within a niche or a genre. We are not opposed to that. We simply want to learn what you want.
Knowledge at Wharton: What is the incentive for the user — the student in this case — to opt in, other than the idea of doing the right thing?
Griffin: That is always going to be one of them. And a better service.
Knowledge at Wharton: Are you also offering to get them out from under the threat of litigation?
Griffin: For us, that has nothing to do with it. In fact, I hope that’s not a factor because I don’t think they are suing people any more. And I certainly don’t think that’s [happening] on college campuses the way it used to. I don’t know because I’m not involved with that part. And we never were. Some people say, “Oh, well, this is the quid pro quo for that” –which is not accurate. Because we were doing this before they made any decisions with regard to that.
Of course, that has to be a backdrop, yet it’s not interesting to us because we are interested in what persuades people to pay — not what compels them to do so.
We think about how can we make the service better? We have a critical issue that is those who offer all music can do so because they license no music. In other words, piracy is one of the factors we are competing with.
We think we are competing more with a limited clock and a limited wallet. We don’t think our competition is pirates even though they are part of the competition. The biggest part of the competition is that there are other things for you to spend your money on and other ways for you to spend your time on other than music. And while music might be a part of your spending your time — there is music in video games and in movies. When I was a kid there were no cell phones, so I wasn’t spending money on cell phones. I wasn’t spending money on network services or electronic games. How do we compete for that share of the wallet that we used to have? How do we get your attention and have you spend the time on your limited clock listening to music and paying for music? And then, of course, there is piracy.
And so we think of that as a holistic problem, and we think, “How can we make it a better experience?” I spend a lot of time [on piratical networks] because I’m really interested in what people do and what they use — I have found that my experience is less than satisfying.
Knowledge at Wharton: Does the RIAA know about this? [Laughs.]
Griffin: Sure they do! And I think to their credit they are beginning to learn in checking that out, too.
Peered networks have a real advantage in their distribution of music — because it can arrive faster. It could be that it’s the person in the dorm room a floor below you that has the song you are looking for. And it will arrive faster if we pair you with that individual than if we sent it to you from a centralized server. And you want fast. You want efficient.
Knowledge at Wharton: What do you think is the likelihood that all these pieces will come together and we will get to a point where artists get their fair share of the payment, and users can download any song they want for a modest fee?
Griffin: I think it is inevitable. The question is “When?” We do it now for radio and for television. We do it for pub owners. It does work. They can play any music they want and they are not considered pirates or thieves.
It is a shift for rights holders to think in this new way. But for users the shift has already occurred. They already get access to all the music they want. They already get to use it with great facility. And I think they are willing to pay, but we haven’t offered them the kind of model that is attractive to them. No one has ever offered an unlimited download, unlimited access model, with one flat fee without any DRM, without any controls at all. They’ve never been offered the thing that they are most likely to want to use.