February 12, 2003
Outline of Professor Fisher’s Speech at Stanford’s CIS
On January 13, 2003, Professor Terry Fisher spoke at the Stanford CIS about his proposed plan for copyright. Listen to the webcast here. I’ve taken these notes – hope they’re helpful.
1. Potential advantages through fuller exploitation of technology
Distribution today
Sound recordings done through:
a) CDs
Standard money distributed for typical CD purchase: 35% retailer, 27% record company, 15% manufacturer, 9% distributor, 16-17% for artist (that latter number is a ceiling)
b) Broadcasts, which are promotional
Fisher notes payola, though illegal, is still practiced
Movies
“Temporal price discrimination” – releasing movies in a series of formats over time, trying to milk most lucrative markets. Done to maximize revenue (“second degree price discrimination”)
How can the Internet change distribution, thus making world better?
Methods: Downloading, “Interactive” or “on-demand” streaming (eg, pay-per-view) (choice), “Non-interactive” streaming (eg, webcasting) (no specific choices)
Improvements:
1. Cost savings (see CDs above) – eliminates need for retailer, some of the functions of record company – you still need a distributor, but probably a less expensive one. Roughly speaking, 2/3 of costs have disappeared.
2. Eliminates over/under-production – copies are “produced” on demand.
3. Convenience and precision are good for consumers. Idea of “celestial jukebox” – being able to pull all types of media down from Internet at a whim.
4. Increases number, variety of musicians because of reduced costs. Also, new distribution mechanisms are creating niche markets, increased demand for marginalized music.
5. Increase “Semiotic democracy” – the reversal of trend in Western economies in 20 century towards concentration of who makes “cultural meaning” – fewer music, movies, etc. companies. Semiotic democracy improved by new tools to manipulate art to rip, mix, and burn; make new materials, and shape how materials are presented to you (think CleanFlicks and ReplayTV). Blurring boundaries between producers and consumers.
Disadvantages
1. Piracy
2. Protecting “integrity” of works – integrity of artists
3. If there is no recognizable version of particular movie, how can we have that shared experience around art?
2. What’s Gone Awry (skims through it)
Something promises happens, things fall apart. Why do things fall apart? Organizations that will lose fight back.
Fisher discusses everything from webcasting, to Napster and Scour, to CD burning, to KaZaA.
Why is this bad:
1. High transaction costs still.
2. No celestial jukebox.
3. Encryption (DRM) and ephemeral downloads and such impede fair uses and semiotic democracy.
4. Continued concentration of industries reduces consumers choices.
5. Piracy hasn’t stopped.
6. Notes independent record stores (HMV and Tower, not the Walmarts) are most hurt. Walmart only features certain types of music = bad for choice.
3. Alternative Legal/Business Regime
A. Taking Property Rights Seriously.
Make copyrights mimic property law (Centrality of owners right to exclude, right to injunctive relief, right to secure aid of court to prevent infringement, civil remedies with criminal sanctions, prohibition of circumvention tools, privilege of self-help)
Privilege of self-help is limited – can only use it when you legitimately have right to exclude and police would do it for you if you ask (not all situations. Ex. someone else’s boat tied to your dock, you can’t just untie it – that’s illegal, even though the dock is your property)
Policies:
Full public performance right, not qualified by compulsory licenses.
Sound recordings (record companies) would get revenue from radio stations (don’t now – only songwriters, composers and publishers get it).
Purge all compulsory licenses.
Enforce NET Act for criminal prosecutions.
Go after SuperNodes on KaZaA.
Hollings bill, requiring copy protection recognition in all devices.
Curtail anti-circumvention provisions on DMCA and impose liability for using technology to impair public’s rights in copyright.
Why? If we’re aligning copyright with property, have to align idea of self-help. We should respect exceptions to rights to exclude and avoid technical arms races
Benefits:
1. Would accelerate Internet distribution by media companies. Criminal application of laws would reduce free alternatives. Less KaZaAs, more MusicNets and PressPlays.
2. Expansion of royalty collection societies like ASCAP and BMI – those societies have been good at licensing, more flexible and able to meet different industries needs. (Cites Professor Robert Merges)
3. More precise signaling by consumers – precisely see what they want, what the demand is
4. Easier for musicians to break free of record companies. In this environment, musicians could make money if they were protected by substantial degree from erosion of their markets through piracy. Enabled to make money on their own, musicians would flee record companies.
Disadvantages:
1. Chilling many critical uses of art. Copyright owners will engage in more price discrimination enabled by DRM to fine tune conditions consumers get access to work (take advantage of willingness to pay – fees related to access, use). Suppliers could charge more money for making uses of products (ex. people dependent on sampling)
2. Erosion of end-to-end principle of work. Internet built-on multipurpose devices at edges of system. Cable providers could control content of material that moves through. Standardization/dumbing down of computers on the end. Limits flexibility of technology. What uses would be blocked? We don’t know – that’s the problem.
B. Compulsory license scheme
Treat entertainment industry like a public utility. Industries of that sort have been treated differently. Arguably, entertainment industry resembles electric company rather than a distributor of real property. It invites that treatment – oligopolistic, should calibrate to maximize public interest, characterized by unequal power between firms rather than consumers or creators.
Increase state intervention. This solution contemplates precondition preserving DMCA and prohibition on file sharing.
Government would regulate prices, establish categories of Internet distribution, setting terms and conditions, setting the rates. Would determine who gets what shares (record label or artists or publishers or composer?)
Benefits:
1. More internet distribution.
2. Reverse “vertical integration” because any intermediary could get permission to distribute. That means new business models.
3. Revenues of record industry would remain stable.
4. Composers and performers would get larger shares.
Disadvantages
1. Large transaction costs. 25 million to make fees for non-interactive webcasting – exceeds amount that would be processed through it.
2. Curtail semiotic democracy – still have copy protection.
3. To prevent circumvention, you’d expect privacy invasions. Surveillance of use.
C. Fisher’s proposal
Treat it like a public good. You could have government provide to get around underproduction (because of free rider problem.) Or, you can subsidize some private group to do it. Or, reward afterwards (like Nobel Prizes). Or, can do stuff like trade secrets. Or, can give creator a monopoly.
Better to provide for others to create, reward.
How to do it:
1. Mark – embed watermark. No incentive to strip it out.
2. Tax – tax things that are used to consume media. Fisher says best thing to tax is not the gizmos, but rather ISP access.
3. Count – private agency or government would estimate how much each thing is being consumed. ASCAP, BMI already do this sort of thing. For streaming, webcasting would keep logs. For downloading, suppliers supply data. For CD copying, you extrapolate from CD sales. P2P is most difficult – require P2P services to supply data and then supplement with sampling via the router or ISP level (put people beside the highway, and have them write down the license plate numbers – don’t know where they’re coming from, or where they’re going to, just know that they’re going).
4. Pay out – distribute revenue given frequency of consumption.
Then, lift copyright law completely – no ban on derivative works, copying, translation, etc.
Rough numbers.
Music industry of American composers and assignees is 1.7 billion dollars (note: this is extrapolation of data about a year old). Recording industry is 13 billion dollars a year. Remember, though, we’re saving about 2/3 via new technology (see above). So, the number we’re worried about isn’t 13 billion but, say, 6.5. Say we try to replace this money. 150 million people who have access to Internet in US, on 50 million dial-up and 12 mil broadband. Former is 20, latter is 44 per month. Tax each 11 dollars per year or .94 cents per month, so impose 2 dollars on each dial up and 3.83 on each broadband. Why different prices (proportionally)? Price discrimination to move towards more socially beneficial access.
Merits:
1. Large cost savings.
2. Lower transaction costs (compared to alternatives).
3. Elimination of dead weight loss. Under digital copyright regime, marginal cost = 0. We want people to consume at that level. Marginal revenue = marginal cost leads to optimization. People eat until they’re full of media.
4. Creators would be paid in proportion to demand. Emphasizes that this is not socialism. This regime mimics current regime in that consumed demand is determinant.
5. Convenient, easy.
6. Averts concentration of industries. Anyone can do this; artists incentive is to maximize distribution.
7. Semiotic democracy – materials would not be encrypted.
Impediments
1. Risk of ballot stuffing. Incentive for creators to exaggerate frequency. Would have to adjust sampling to deal with this.
2. Some threat to privacy – maybe people could use watermark to see who’s using what. Not essential to system though.
3. International problems – how could you get everyone to adopt it?
Audience questions/reactions (this is a bit harder to scribe)
Fisher wants to start small with music industry. Understands that if this takes hold, the tax numbers might have to go up. Also, should start with generous pay outs to creators, to urge adoption.
Fisher notes that in the end taxes paid won’t have to go up too much, because music industry might be over paid at this point. Just need to compensate them fairly.
Why no need to strip watermarks? It’s a taxed based regime – there’s no choice whether you pay it. No incentive to circulate non-watermarked files. This would be a problem if you strip out someone’s watermark, put yours in. Fisher still needs to work out how to sanction people who do tamper/ballot stuff.
(Can’t hear this well, but) What if song included in a film? Fisher isn’t sure yet.
First, what about people on wireless networks beneath ISP level, or off-site supplier? Second, can you extend it to all digital media? As to second, Fisher says yes. As to first, you still have to pay local tax, and if media came from American artist, you’ve still got the watermarking. But, if we have a wireless commons, we would have to shift structure, so you might have to take the money out of general tax revenues.
What will happen when system expands? Fisher says prices will go up, but will be mitigated hard to document cost savings. We’d also have to constantly adjust total amount and definition of categories of works (have to differentiate between what each type of media deserves). Going global would help.
How has this been tested out in practice? Well, not the sampling. But he has gotten some good responses from rights collection societies and from head of SoundExchange.
What would be the new incarnation of record company? Fisher notes how it’s much easier to produce music, and that the A&R (scouting) function might not be necessary. Promotion is more tricky, but can be disaggregated. Also, if artists need their sound to be more popular, they could go right to people who help with that rather than record companies.
Fisher notes how vertical integration is dangerous, has been so in CARP proceedings.
Are you concerned about limiting broadband growth? Fisher thinks it will increase growth, more attractive product.
Filed by Derek Slater at 10:07 pm under
Comments Off on Outline of Professor Fisher’s Speech at Stanford’s CIS
