The Economic Impossibility of Copyright in a Digital World

I recently posted an article (A Conversational Treatise on the Problems with Intellectual Property) that addressed what I believe to be the incorrect assumptions necessary to support the entire concept and legal system surrounding intellectual property. The post came about as a result of a conversation I had with a friend, and it is simply a question/response format long discussion on the topic. The core argument I made is that information is inherently not ownable, in that it is an infinitely reproducible non-physical resource. Secondarily, patents and copyrights are not necessary for innovation and creation to flourish.

My friend asked good questions throughout the course of the conversion, which helped me articulate and solidify my own viewpoint. However, I admit that some (though not all) of that conversation and the arguments within it are ideological in nature, and based on my worldview rather than incontrovertible empirical data. Therefore, I want to address a different side of the argument, one which is based on current and widely accepted economics.

I know that my ideas here are very, very far away from what many people believe, so I don’t expect everyone to agree with me. All logical comments in response are welcome. The core of my argument today is this:

Digital storage and distribution, combined with basic economic principles, are fundamentally incompatible with selling zero-value-added digital information.

In other words, “create once, sell forever” is no longer a viable business model if all you deal with is digital information.

Ideological Argument Recap

One of the comments on my last post presented an example scenario to counter what I wrote there. The essence of the scenario is that someone decides to take everything I wrote in the article, claim it as his own, and then use it to get a book deal with the Wall Street Journal. I’ve just been ripped off, haven’t I?

In short, no.

I will recap my original argument here for clarity: in this case, the knowledge has already been given away (I posted it online), and to demand future payment for that is unjustified. I might be totally happy not to pursue the topic any further, and then the rest of society might benefit from someone else continuing the discussion via a WSJ-sponsored book. But assuming I did want my own book deal, then as the original author, I also have means of attempting to expose the other person as a fraud, redirecting the WSJ to the original source of talent (assuming they consider it a “talented” composition). But the only thing really wrong in this hypothetical situation is that the other guy lied about his authorship of the original work—a problem which does not require copyright law to address. The fact that he got a book deal based on what I did means that I am passing up a potential opportunity, not that he “stole” the published content that was “rightfully mine to sell.” He does, after all, still have to do all the work of writing the book in order to fulfill his end of the WSJ deal.

The same friend I had the original conversation with asked a similar hypothetical question, this time about the “S#!@ My Dad Says” guy on Twitter. He started a while back posting short, funny quotes from his dad on a Twitter account, and as a result of that, ended up with a book deal and a TV show. (Good for him for capitalizing on it!) My friend’s question was, like the post comment, what if someone else had compiled all those tweets and printed a book or started a TV show without accrediting the original author? Wouldn’t that be screwing him?

Again, in short, no.

It would still be wrong for another person to claim authorship of something he didn’t write, and it would be an obvious move for the original author to jump in and demonstrate his true authorship, and subsequently reap the benefits of whatever negotiations had been fraudulently started by the other guy. However, that’s not a problem that requires copyright law to address. It’s just fraud, plain and simple. If the other guy managed to get a book deal or a TV show based on stuff that admittedly belonged to someone else, then there should be no problem whatsoever.

The Economic Reality of Marginal Cost

Now, to address today’s topic: Whether or not you agree with the previous argument, there is an economic argument as well which starts the same way. Exactly why is it actually wrong to “monetize off of somebody else” in this way? Fundamentally, it’s just duplicating information and then adding value (compiling it into a book, etc.) which others are willing to pay for. The original author, by virtue of being the true source, has options for attempting to redirect the money back to himself as I mentioned before. But by freely posting his content online, he has permanently set the no-value-added “cost” of that knowledge at exactly $0.

If the SMDS guy didn’t make a book or TV show out of his tweets, and nobody else did either, then everyone would be content (with the possible exception of the people who really wish there were book or TV show). Nothing would have been “taken” from anyone despite the fact that it is all freely available for everyone to read. But suddenly, once a book or TV show deal comes into play, it makes a legal difference who goes forward with it? That doesn’t make sense.

The economics of this are simple. First, a few definitions courtesy of Wikipedia’s pages on marginal cost and marginal revenue.

1. Marginal cost is the change in total cost that arises when the quantity produced changes by one unit.
2. Marginal revenue is the extra revenue that an additional unit of product will bring. It is the additional income from selling one more unit of a good; sometimes equal to price.
3. For a business, profit maximization requires that a firm produce where marginal revenue equals marginal costs.

Note here that the marginal cost of information (knowledge, ideas, art, music, video, designs, etc.) is effectively zero. I can share information for free. I can create 1,000 copies of one ebook, MP3, or video as easily as 100, or 10, or 1—or none, for that matter. Anyone else can too, just as easily. As storage and bandwidth costs continue to fall, the marginal cost will only get closer to zero—though it is already so close as to be indistinguishable. Economically, this means that sustainable marginal revenue (i.e. markup) can only be greater than zero if there is something besides the information itself present to add value. This is why it is impossible to enforce copyright today. Attempting to do so is directly attacking a fundamental principle of basic economics. It is not possible, in the long term, to make any money on something that costs nothing to reproduce unless you add something else of value along with it.

Although it is painful to face sometimes, this applies to any information that becomes public, whether intentionally or not. The effective economic cost of any public information is $0. This is why trade secrets are valuable, and yet even our IP-friendly government says you’re out of luck once they become public. They are no longer secret, and impossible to protect as such.

Real Work Creates Real Value, But Not Necessarily Forever

Now, if you’re in a band that goes to the effort to record a dozen songs for an album, that work is real and does have a cost. The typical thinking today is that the cost of that work should be off-loaded onto the consumer in exchange for albums, concert tickets, merchandise, etc. This is a fine idea except that album sales, economically speaking, should not be included in the revenue equation due to their $0 marginal cost. Whatever else it may be, relying on album sales for a critical stream of revenue is an unwise business model, because it depends on your ability to convince people that the value of a $0 marginal cost resource is not $0. There are plenty of less risky ways to offset the fixed cost of the original recording, most of which are some private or crowdsourced variation of the concept of commissioning a work of art.

I also want to clarify that while the digital data representing a band’s album has a $0 marginal cost, that doesn’t mean that it is impossible to make money selling albums. Obviously, physical media has a cost, and you can get some small profit by marking up and selling CDs to people who still want physical media. That customer is evidently a dying breed, however, and I wouldn’t recommend it as a key part of your business. A more effective approach would be to sell autographed CDs, or CDs that come with a unique photo of the band, or something like that. These add-on features provide real value by virtue of being unique (scarce, not infinite). Yes, it takes more work than simply running off 10,000 identical copies of a CD. But that’s kind of the point: it takes real effort to create real value. To believe otherwise is naive. The initial work that goes into recording an album does not somehow translate into infinite future value, which is precisely what copyright would have us believe. It is just an excellent way to promote your band and entice people to pay for scarce goods, like concert tickets. But the music alone, in digital form, costs nothing to reproduce and has no defensible inherent value.

Of course, there are a lot of people who will still pay $1 per track from iTunes even though they could torrent your entire discography for free in 20 minutes. Some do this because it’s convenient (or they don’t know how to use BitTorrent). Some do it because they want to support the artists—though I’d argue that torrenting the music and sending a $15 check straight to the artist would probably be far more beneficial. Some do it because they are afraid of the legal consequences. All of these people see a $0 marginal cost product as having a value above $0, due to altruism, convenience, or fear. The music industry has the best chance to succeed by capitalizing on convenience, but they seem to run in the opposite direction all the time by employing ever-worsening DRM and by attacking services that make it easy to get lots of music while still paying, simply because it’s less money than they demand.

DRM Has a Very Temporary Value

I am not implying that anyone who produces a digital product—music, movies, photos, or a software package like Windows or Photoshop—should by definition give it away, either in compiled or source form. It is certainly up to the creators to do whatever they want to with their products, including the application of complicated DRM and copy protection methods. Economically, this adds a perceived “value” of convenience that (to some) makes it worth paying for the official version, because it means they don’t have to do any extra work to get around any DRM and copy protection.

But this is a temporary solution, since DRM can always be broken with enough effort, and then the “value” is effectively gone. As content creators get more and more desperate to protect their content by employing these methods, they often end up annoying or completely alienating their paying customers while the pirates continue to have an enjoyable, DRM-free experience. In some cases, the copy protection schemes are so bad that companies are forced to issue patches to defeat their own DRM, as was the case with the infamous Sony rootkit debacle, and with Ubisoft—which actually turned a 3rd-party software crack into an official utility to “fix” one of their games.

Copy protection measures are a gamble at best. It takes more work to implement, more work to update, more work to support, and for at least a small portion of your customer base, it will end up being completely worthless. This is true of video, audio, text, and software; anything digital is subject to this reality. Of course, it is definitely your prerogative as a creator to use DRM if you want to, but it’s a big risk. Economically, it makes more sense to focus on adding real value that nobody can take away from your product by breaking through the DRM. Depending on the nature of your product, examples of real value are things like priority and/or automatic updates, helpful customer support, custom integration, autographs, concerts, merchandise, personal communication, consulting, informational workshops, training sessions, and media appearances. Anything that requires actual effort every time has the potential to create real value, since actual effort cannot be duplicated at zero cost.

Create Once, Sell…Forever?

Back to the “S#!@ My Dad Says” guy on Twitter: assuming that only the tweets exist, then the only way anyone (including the author) could sell a book or create a TV show is by putting in more effort. If the original author does it, great. If someone else does it, fine. In an economy that recognizes that the cost of digital reproduction is zero, it shouldn’t make a difference who does it, though with some effort, the original author probably has the best chance to effectively monetize his own work if he wants to. The author has already eaten the fixed costs and set the marginal cost of his creation to $0. To expect automatic remuneration for doing nothing else is an unjustified sense of entitlement, albeit one that has been encouraged by our IP system for hundreds of years.

Why does it make sense for a product to be created once (in an infinite format) and then sold exactly as-is to an unlimited number of customers for a profit without adding any further value? The concept is absurd outside of the information business, since any other type of product cannot be reproduced at zero cost. This, really, is the whole point. It’s only been possible to pull this off inside the information business because there previously weren’t enough tools available to counter it effectively—though, there have been information “pirates” in the classic sense for a very long time. Then along came the internet and digital storage and transfer, and suddenly people everywhere are realizing that a zero-marginal-cost thing with no value added should maybe also really be a zero-marginal-revenue thing.

People tend to complain when markup is very high, because it feels like the merchant is trying to rip them off. Markup is often calculated based on cost, which is done by dividing the difference in cost and sale price by the original cost. In the case of infinitely reproducible zero-marginal-cost goods, this means that any markup at all is always infinite. No wonder so many people reject the idea!

The development of cheap digital information storage and zero-cost duplication and transfer has killed the viability of the “create once, sell forever” content-centered business model. Whether it should have is a different discussion, but it certainly did.

The Decadence of Society or the Failure of a Business Model?

“But Jeff,” I hear you say, “are you claiming that just because technology makes it easy to steal, that it’s somehow okay? What have you been smoking? If technology made it easier for you to walk out of my front door with my plasma TV, would that be okay too?!”

Ah, no. That is missing the point entirely.

Physical property is very different from information (a.k.a. “intellectual property”). You cannot own information, and therefore you cannot steal information. Stealing your plasma TV, on the other hand, would be wrong no matter how easy it is.

Technology has not made it easier to “steal” information from people who “own” it. Technology has simply made it a lot harder to sell something which could not really be owned in the first place. What we are witnessing today is a violent backlash from businesses that previously made all their money by selling identical copies of infinite resources. This business model has been effectively killed, though the law hasn’t caught up with this reality, and those who stand to profit most are fighting like mad against their impending doom (though a few are adjusting their business models instead).

The nominal purpose of the institution of copyright is to encourage creativity. I address at length in my last post why I don’t believe this is necessary at all, but for the next point, its necessity is irrelevant. Copyright is enforced by establishing a specific owner of some piece of information, and then that owner can file a copyright infringement lawsuit against someone else if they use that information in an unauthorized way without permission. Of course, even if the lawsuit is won, the defendant can’t exactly give the information back, since it was never gone in the first place, but…I digress.

The point is that copyright law is based on the assumption that information can be owned, and subsequently enforced by attempting to keep people from stealing it. Since it is impossible to deny that information is itself infinitely reproducible at no cost, the only practical purpose for copyright law is to make sure that information (with no added value) is sold for some monetary gain according to the desires of the copyright owner. In other words, the only practical purpose for copyright is to prop up a business model that is economically unsustainable.

Again, whether it should be is another conversation, but there it is.

One other clarification: my ideas don’t change the fact that IP laws are still in place. No matter how much you or I or anyone may hold this position, you can still be sued for copyright infringement. Ideological difference is not a valid defense in this case. Please keep that in mind, and realize I am not encouraging anyone to break the law. But I wholeheartedly encourage anyone and everyone to create products and run businesses without relying on or enforcing any part of the concept of IP. I think this is the most effective way to bring about change, because it demonstrates that IP is not necessary.

The internet has illuminated the economic problems with IP. Once enough people realize that they can do far better in the long run focusing on the economic solutions instead of IP enforcement, the laws will lose favor and, I believe, ultimately be widely ignored and eventually changed or repealed as a result. One can hope so, anyway.

Alternative Business Models

So if you can’t continue to sell infinitely reproducible zero-value-added information, how can you stay in business? If you’re honestly asking that question at this point in time, then you have a fighting chance.

The obvious (but general) answer is that you need to add some value to the infinite digital product. If people don’t see value in what you are selling, they won’t pay. Costs are measurable, but perceived value is subjective. If the perceived value equals or exceeds the price, then people will pay for it.

Pay What You Want

As I mentioned before, some people will pay even when they know the marginal cost is zero. You can market a product with this in mind using what is typically known as a “pay-what-you-want” model. The Humble Indie Bundle software sales have done this with great success in the recent past. Radiohead was famously one of the first major bands to abandon the common recording contract model and release a pay-what-you-want album (that link goes to a Time article from October 2007 with some very interesting observations). Many open-source projects get some financial support using the same concept.

Pay-what-you-want can work very well to build interest, and it’s almost impossible to lose if your product is digital. But I have a feeling that it only works as well as it does now because people still expect to have to pay for digital information in most cases. If that changes, then pay-what-you-want will lose some of its novelty, but it can still be very useful.

Pay-what-you-want also makes it incredibly easy to go through official channels to get a digital product and effectively donate what you want to. Many people realize that even if they can get something for free, it is still worthwhile to support the creator to help ensure that he continues updating or maintaining that product, or will decide to create another desirable thing in the future. People associate quality with value, and if they get Product A for free and find it to be a high-quality item, they will be much more likely to pay in advance to encourage the creation of another high-quality Product B. Essentially, the creator is saying, “I’m giving you Product A for free. If you like it and you ever want to see a Product B, you have to give me some money, because it’s not worth it to me otherwise.” How much money the creator needs is up to him, and how much money the Product A consumers are willing to give is up to them. If the creator gets enough money to make it worth moving forward to him, then the business continues. Otherwise, it doesn’t.

Of course, this approach is not the only option. It is very difficult to predict how well it will do, so if you’re creating a solid business plan to bring in a reliable revenue stream, you will probably want to incorporate other means instead of or in addition to a pay-what-you-want offer. The initial work (fixed costs) required to produce the first copy of a photograph, book, song, movie, etc. is significant, and it’s good business to try to recoup those costs somehow.

As with any business, the initial investment required to produce the first instance of a product is usually larger than that required to produce the second instance of the same product. With manufactured physical goods, this usually looks like a 1/x curve that starts at one point then drops toward zero. The highest point is usually at the “Quantity = 1” point at the very beginning, and this represents the fixed cost: manufacturing tools, preparation, materials, business licenses, and everything else you need to get started. The lowest point the curve ever reaches is the marginal cost of the product.

In the case of company who wants to sell their products digitally, this is not a curve at all. The “Quantity = 1” point represents the cost of creating the product (which can be very high), and then every point after that is on the zero line. Aside from marketing and distribution channels that require raw materials (printed books, CDs, DVDs, theaters, etc.), all of the costs are at the very beginning of the production cycle. In light of the previous discussion of marginal cost relating to marginal revenue, relying on digital sales of a zero-value-added infinite product to recoup the initial investment is an economically bad move.

Crowdsourced Funding

What other choices do you have, since pay-what-you-want is unpredictable and selling infinite goods is bad economics? Regardless of what your product is, there is always the option of the “commission” approach. By this, I mean that someone else pays you to create what they want, rather than paying you after they discover that you created something they like. The idea of “commissioning a work of art” may bring up mental images of royalty negotiating with Mozart, but nothing says it needs to be complicated or high-profile.

One great example of a current similar system is the Kickstarter platform, which allows for crowdsourced funding with minimized risk. You pitch your idea as a Kickstarter project and set a minimum funding goal, and then if people like it enough, they pledge to support you in return for various “rewards” which you define. If you get enough pledges to reach your funding goal, all of your backers’ credit cards are charged for their pledge amounts, and you get the money you need to continue. If you don’t reach the goal, then your backers pay nothing and you get nothing; if this happens, then the market has spoken, and it doesn’t like your idea enough. Back to the drawing board.

There are countless creators using Kickstarter in this way right now to fund photography projects, books, music, movies, and all kinds of other things. The idea is economically solid and obviously effective. The beauty of this platform is that not only can you raise all the money you need, you also have an excellent opportunity to uniquely connect with your best customer base and get great feedback before you put in all the effort. On the other hand, if your idea has difficulty finding a market, then you get to make that critical discovery before you put in the initial investment to create a product.

The Non-Infinite Component

If you want to sell digital data with any real success, then you need to add some special, unique, non-infinite component to go along with it. (I use the term “non-infinite” rather than just saying “finite” to highlight the fact that the core product is actually infinite and has no marginal cost.) As I listed before, this could take any number of forms depending on the nature of your creative work.

• For software, it might be customer support, automatic updates, inclusion in beta tests, customization, integration, or training.
• For art, it might be a personal showing, unique photos taken or paintings made specifically for the customer, or signed works.
• For books, it might be an autographed copy, a one-on-one session to discuss the book, or a presentation at a conference.
• For music, it might be concert tickets, small-venue personal shows, autographed photos or CDs, T-shirts or other clothing, featured remixes, or back-stage passes.
• For movies, it might be theater tickets or coupons for free food at the theater, a unique single-frame printed photo from the film, signed memorabilia from the set or related to the actors, or other movie-related merchandise.

These ideas are not original to me, and this is not a comprehensive list. The important thing is to make sure people have a reason to buy rather than simply obtain what can be infinitely copied without loss of quality. All of the above suggestions are impossible to duplicate without cost, which is why they work. Sell the things that are finite. Don’t try to sell the things that are infinite, because people are starting to realize that to do so makes no economic sense, whether or not they can articulate it.

The High-Dollar Movie Business

“But Jeff,” I hear you say again, “Avatar cost $236 million to make. Do you seriously think they could have used crowdsourced funding to achieve that, or relied on merchandise sales or other gimmicks?”

In all honesty, probably not. It did gross over $2.7 billion though, a record-setting amount, so obviously a lot of people liked it. The important thing to mention here is that $2.7 billion came in from theater releases. As of the end of July 2011, Avatar only brought in a paltry $190 million from DVD sales, or 7% of the income from the theatrical release. Blu-ray sales are not included in this amount, since I haven’t been able to find accurate information. Taking into account the higher cost and currently smaller market for the Blu-ray format, let’s assume for the sake of argument that Blu-ray sales brought in another $200 million (though it may be much less). This means that Avatar managed to make over $2 billion from theaters in the first two months, whereas 15 months of home media sales brought in only about 20% of this amount. Clearly, theaters offer better return-on-investment opportunities for expensive, visually impressive films.

For movies with very high production costs and a lot of publicity, their best chance to succeed comes from capitalizing on the unique, finite aspects of movies—most notably, as mentioned above, from the theater experience. This experience cannot be duplicated without cost. It’s a social event, something which most people can’t even begin to imitate in their own homes. Today, going to the movies is less “fun” to many people than it used to be, particularly to people who have memories of what it was like many decades ago. But this is not a terminal problem; it is just an area for improvement. Theaters obviously still bring in a lot of money despite whatever shortcomings they may have.

If a private company has a reasonable expectation that a movie will cost $250 million to produce and will gross $2 billion from the theatrical release, then it is a wise business decision to simply get a $250 million loan and go ahead with it. This is an example of a good risk, and a good use of debt, to fund an expensive creative undertaking. No copyright is necessary to create the story, hire actors, hire a fantastic digital effects crew, turn it into a visually stunning movie, or get a 700% return-on-investment.

Game Theory and Copyright-Free Content

For the sake of thoroughness, there’s a counter argument here that I need to address: if the marginal cost of a digital copy of Avatar is $0 (which it is), what’s to stop the theaters themselves from “pirating” the $236M-fixed-cost movie in a copyright-free world and just reaping all the $2.7B profits for themselves?

This is a very good question, which does indeed have a good answer consistent with a copyright-free position.

Movie theaters are nothing without movies. They cannot possibly survive if they don’t have good films to show on a regular basis. Therefore, the first priority of a theater that wants to compete is to make sure it can obtain new movies in a timely fashion. It can get them through an authorized contract agreement from the official distributors, or it can attempt to get copies for free through unofficial distributors. Any smart theater owner is going to go the free route, right?

Actually, no.

Game theory indicates that businesses in a competitive market will almost always choose the option that has the most “predictably good” outcome, not the option with the highest possible profit, when the outcome depends on the decisions of competitors. For instance, let’s say that two people can each choose either option 1 or option 2. If they both choose option 1, then they both get $5. If only one of them chooses option 2, then that one gets $10 and the other one gets $3. But if they both choose option 2, then they both get nothing. The only way either can get $10 is if only that one chooses option 2. Although it is possible to get $10 with the right set of choices, the best predictably good choice is option 1, which has a smaller guaranteed payout, but no chance of a zero payout.

For a movie theater, competition means showing new, popular movies in a fun environment and at a reasonable price as soon as they are available. If the owner opts to show movies obtained through unofficial channels, he is likely to run behind the theaters who don’t, giving them a clear advantage among people who want to see the movies first. This decision also virtually guarantees that no other major movie production company will make a deal with him in the future, meaning that theater is forever relegated to showing unofficially distributed movies behind its competitors. Some “bargain” theaters would be able to make a business out of this, to be sure, but not all of them.

Bringing the game theory example into it, the theaters who use official channels are choosing option 1, the smaller guaranteed payout instead of maximum possible profit. By doing so, they are also fostering a relationship with the movie production companies and building goodwill among their providers and customers (who want to see more good movies made and shown in the future). The other theaters that use unofficial channels, on the other hand, are choosing option 2. They get a larger profit per ticket, and by taking some customers away from the official theaters, reduce their competitors’ profits as well, but at what cost? Delayed releases, zero beneficial relationships with the movie producers, no future contract possibilities, and boycotts from customers who believe some of the ticket price should go back to the producers.

Continuing the game theory application, what if all theaters decided to choose option 2 and bypass official distribution agreements? For a very short time, they would all reap enormous profits. Then, of course, their source of profit would vanish entirely. Most people, and certainly most production companies, would never spend $250 million to make a movie which they know they can’t recoup their investment on somehow. Few people would be passionate enough to spend even a tiny fraction of that hiring actors and crew to create a movie if they knew without a doubt that someone else would reap all of the profits on it. Game theory indicates that this “mutually assured destruction” outcome will not happen, but that most theaters will choose to stay on good terms with producers to support what ultimately creates their customer base.

This is not to say that the film industry, or any creative industry for that matter, will die without a profit motive (I totally disagree with this idea). There are many reasons people create every conceivable type of work. For every blockbuster movie, there are probably dozens of excellent small, independent films that never get much exposure. These are often made on a very small budget, sometimes with no expectation of a guaranteed return-on-investment. But for the argument above, blockbuster releases are the lifeblood of many theaters, and it is definitely true that these high-profile, giant-marketing-campaign films are made at least in large part due to their profit potential.

…which, again, is really not dependent on copyright.

Summary

• Ideas cannot be owned.
• Copyright is not necessary for creativity.
• Digital storage and distribution, combined with basic economic principles, are fundamentally incompatible with selling zero-value-added digital information.
• Public information has a permanent no-value-added “cost” of exactly $0.
• Sustainable profit can only be greater than zero if there is something besides the information itself present to add value.
• DRM has a very temporary “value.”
• There are plenty of good ways to recoup initial investment that do not rely on copyright.
• Technology has not made it easier to “steal” information from people who “own” it. Technology has simply made it a lot harder to sell something which could not really be owned in the first place.
• The only practical purpose for copyright is to prop up a business model that is economically unsustainable.
• The development of cheap digital information storage and zero-cost duplication and transfer has killed the viability of the “create once, sell forever” content-centered business model.

Any logical comments are welcome. I think I’m right, but I don’t know I’m right. I’m always open to hear from people who don’t agree.