Excerpt from The Knockoff Economy: Tweakonomics

Here is an excerpt from The Knockoff Economy: How Imitation Sparks Innovation, which has just been published by Oxford University Press. Over the next few weeks, we’ll be running 2 excerpts from the book here on the blog and taking questions from Freakonomics readers in a Q&A. We’ll also run a contest for the wackiest photo of a knockoff item. 

In The Knockoff Economy we examine the relationship between copying and creativity. Most people who study this area look at industries such as music or publishing, where intellectual property (IP) protections are central. We do something different: we explore innovative industries—such as fashion, food, fonts, and finance–in which IP is either unavailable or not effective. In these industries copying is common, yet we find that innovation thrives. In a world in which technology is making copying ever easier, we think these industries have a lot to teach us. And one of the key lessons is that copying is not just a destructive force; it can also be productive. Harnessing the productive side of copying—the ability to refine, improve, and update existing innovations—is at the heart of this excerpt.


Rules against copying don’t just cover outright imitation. They also address variations: works that use that some portion of another creative work but add in new stuff, and in the process transform the original work. Think of Shepard Fairey’s famous Hope poster of Barack Obama, which took an existing photograph and reworked it into an iconic image: 

Shepard Fairey's Obama "Hope" poster.

It is normally against the law to alter a creative work—what we call in this book tweaking–unless the original creator gives permission. That’s because doing so involves some degree of copying, and copying is what copyright law is meant to stop.

For the last century, however, there has been a different rule for songs. (Actually, to be precise, for musical compositions, as opposed to recordings of those compositions). This special rule gave birth to what today we know as the cover song.

Cover songs are very common, in part because the legal system makes it easy to record a cover. The cover artist must pay something to the original songwriter if she sells recordings of the song. But she doesn’t have to ask permission to cover the song and reinterpret, and tweak it, as she sees fit. In short, she can copy the song, add her own variation, and put it out in the marketplace.

Why does American law contain this unusual exception for cover songs? It stems from an interesting historical accident. In the early days of copyright, the rules about music were straightforward for a simple reason: at the time, there was no way to record music or to mechanically reproduce it. Music existed on paper (in the form of sheet music) and in the air during live performances. Copyright law prevented—at least in theory—the unauthorized copying of sheet music, which was the only copying possible.

That all changed after the Civil War, with the invention of the player piano. By the 1890s player pianos were widely distributed in the United States. (The phonograph was invented at about the same time and it too was everywhere by early 20th century.) The player piano deeply troubled popular music composers such as John Philip Sousa. Sousa worried that the pianos would kill the public’s demand for sheet music, and sheet music was the source of composers’ copyright royalties. To make matters worse, the player piano companies refused to pay royalties to composers for the songs they put on player piano rolls. These rolls were scrolls of paper with holes punched out in patterns that instructed the piano how to play a particular song. The rolls, argued the player piano companies, did not “copy” the composers’ musical compositions. As a result, they were perfectly legal.

As Sousa knew, that argument was more than a little disingenuous. Sheet music “copies” a song by rendering it into musical notation—symbols on paper that tell a musician how to reproduce the song. In a similar fashion, a player piano roll “copies” a musical composition by rendering it into a different sort of musical notation—holes punched into paper that tell a machine how to reproduce the song. Sheet music and player piano rolls are essentially the same instructions, just written in different languages.

Nonetheless, the Supreme Court, in its 1908 opinion in White-Smith Music Publishing Co. v. Apollo Co., sided with the player piano companies. The Court held that because humans could not read player piano rolls, they were not in fact copies of the musical compositions they encoded. 

The result in White-Smith lasted but a year before it was overturned by Congress. The Copyright Act of 1909 extended the law to cover all “mechanical” reproductions of musical compositions, whether they could be read by human beings or not. With this action, however, Congress mandated that all musical compositions would be subject to what is called a “compulsory license.” In short, since 1909 the copyright law has allowed musicians to copy others’ songs without asking permission, so long as they paid a specified fee to the original songwriter.

Why did Congress create this system of copying? Because they feared the power of the Aeolian Company.

Who? Aeolian is a long-vanished manufacturer of pianos, player pianos, and organs. Aeolian declared bankruptcy and disappeared in 1985. But back in the first decade of the 20th century, when player pianos were the hot new technology, Aeolian was a dominant firm—the Microsoft (or perhaps Google) of its day.

Anticipating that Congress was about to overturn White-Smith and extend copyright to mechanical reproductions of musical compositions, Aeolian had moved swiftly to buy up song rights from musicians and publishing companies so it could copy them onto player piano rolls. Aeolian’s competitors quickly complained to Congress about Aeolion’s attempt to corner the music market. Congress responded with the invention of the cover song rule. The immediate result was that Aeolian’s competitors gained the right to make their own player piano rolls, so long as they paid the fee. That was Congress’s intent: to keep the Aeolian Company from having a monopoly on the then-crucial player piano roll market.

The longer term result was much more significant: because of Aeolian’s dominance of a now-defunct technology, we have a musical culture in America in which musicians are free to tweak songs they like—and they do so with great enthusiasm. Bob Dylan wrote “All Along the Watchtower”; Jimi Hendrix tweaked it into something quite different and, arguably, made a great song even greater. Another 1960’s classic, Van Morrison’s “Gloria,” has been covered by performers including Jimi Hendrix, the Doors, David Bowie, Tom Petty, Bruce Springsteen, Rickie Lee Jones, AC/DC, and Patti Smith, in perhaps the song’s most memorable and inventive reinterpretation. “Gloria” is an enduring song in part because so many legendary musicians have tweaked it. Cat Power
, John Lennon, Willie Nelson, Paul Anka, and many other famous artists have issued albums of nothing but cover songs. All this (legal) tweaking has made our musical culture immeasurably richer.

But has the freedom to tweak others’ songs, in exchange for a very low fee that the original songwriter has no power to override, suppressed the incentive to write new songs? There’s no evidence of that. Indeed, every day we see a continual outpouring of new musical compositions. Tweakers and pioneers co-exist comfortably in the world of music.

It is sometimes hard to tell them apart, in fact. Think for a moment about jazz greats like Charlie Parker and John Coltrane. On one level, they are tweakers—Coltrane’s version of the Rogers and Hammerstein standard “My Favorite Things” recognizably appropriates that song’s famous melody. But if Coltrane starts there, it’s certainly not where he ends up. By the song’s end more than 13 minutes later, Coltrane has tweaked the original melody and taken it in a much darker, more contemplative direction. At some point in the song, Coltrane crosses the uncertain border that separates Tweakers from Pioneers.

Tweaking is present in all inventive fields, and in some—like music—is a very prominent part of the creative process. Perhaps the most important point about tweaking is this: tweaking does not appear to suppress pioneering innovation very much. If anything, it may often encourage it. Many of the most significant and enduring innovations rest on tweaking. As Malcolm Gladwell has argued, the late Steve Jobs of Apple—an icon of our innovation economy if there ever was one, and the man behind the iPhone and iPad—“was repeatedly referred to as a large-scale visionary and inventor.” But in fact, “he was much more a tweaker.”

Steve Jobs, Gladwell goes on to argue, was “the greatest tweaker of his generation.”Even the iPad, Jobs’ last great success, was a tweak of an idea out of Microsoft.

Eric M. Jones.

Early US copyrights were copies of the European laws (were the laws copyrighted?) and 28 years was about all you got. In 2008 the Sonny Bono Copyright exptension to about 110 years.

These extensions were pretty smooth over time. See: http://en.wikipedia.org/wiki/File:Copyright_term.svg

Subject for discussion: "Should US copyright terms be getting shorter instead of longer?" When the 1790 copyright act was passed, 28 years was at least in the same order-of-magnitude as the time it took for the work to spread across the globe, and most copies took great skill to make. Now a song or book can be all over the planet in mere seconds while the US copyright lasts 110 years, and copies are made at the speed of light by anyone.


Such works can be all over the planet in seconds, but are they? What of the numerous works that only become popular years or decades after their creation? Or what of the backlist, when you discover a work by a new-to-you author/performer, and find that s/he has written a number of previous works? Should you be able to copy them for free?


Copyright gives and copyright takes away. It gives incentive to the person wondering whether to continue their career as a novelist or to give it up to become a real estate agent. It takes away the ability of people to tweak older works.

At what point do the benefits become less than the drawbacks? How much extra incentive do creators get from long copyright terms? Did anyone under a 28 year copyright term contemplate a blank sheet of paper and decide not to start writing a novel because they'd only get 28 years of income from it? I can imagine this would happen occasionally.

Would anyone not write a novel because it would only produce 50 years of income rather than 110? I really doubt it. A very small proportion of works produce significant income over 50 years after being written, and even for those which do, the net present value (at time of writing) of that future income stream is minute compared to the near term income.

For me, publication+50 years seems about right for copyright. It will seldom predecease the author, so it provides as much incentive as today's inflated terms, but allows creative reuse on a reasonable time frame. Life plus 70 years is just ridiculous.



I find that what you are talking about is very similar to a case I've been vaguely following myself. I think that given what it covers, it could change the way IP works.



Every new idea is built on old ideas. Every new thing is a combination of existing things. Our current copyright and patent laws do not encourage innovation, they strangle it. Many of the basic discoveries and inventions that companies across the globe rely on would still be under patent or copyright if the inventors had chosen to protect them. Its a good thing for society that they did not.

A good Ted talk about this.


I think that, philosophically, patent and copyright law is closely related to the Great Man Theory. Basically, the idea is that a few 'great' people determine the course of history and society. If this is true, then it means that innovation is a result of these 'great' people, and they should be rewarded by society appropriately.

Fortunately, the idea of the 'great man' has been thoroughly debunked. Unfortunately, it has been integrated into our laws and how we think about progress. The reality is, if Einstein hadn't invented relativity, someone else would have; the same is true of every significant advance. These are much more the products of society than of any one individual, yet the individual is the one rewarded at the expense of society. If Einstein had been born in even slightly different circumstances its very likely that he would have accomplished nothing of note.


The flaw in this argument that "someone else will discover things" is that it only applies to the natural world. Yes, someone else would discover relativity if Einstein didn't, because it is part & parcel of the way the world works; someone else would have discovered America if Columbus hadn't (in fact, several someones discovered it before he did), because the only way not to discover it was not to make long westward ocean voyages.

However, no one else would have written Shakespeare's plays or Beethoven's symphonies. Other works, perhaps as good, but though there were many playwrights in Elizabethan England, only a couple of Marlowe's plays are even remembered except by scholars.


As an aside, both Shakespeare and Beethoven lived in a time without the modern concept of copyright. This proves that they did not create anything of value. After all, without copyright there can be no creativity or art.

In fact, looking into it it seems that both ruthlessly *stole* from the works of others. Even if the law back then let them get away with it, morally speaking they should have been imprisoned - or at the very least sentenced to lifelong debt slavery - for their thieving ways.

Mike B

Intellectual property protections are important in areas where you have high capitol investment costs because capitol will always flow to the areas with greatest return and few would invest billions to develop, lets say, a new drug, if the price would immediately drop to marginal cost of production.

On the other hand intellectual property protections are far less important where the costs involved are simply that of one's time. People are far more likely to provide free labour than free capitol and in many cases they can even receive non-monetary from their free labour. The capitol costs of producing a new work of music is very low so it is likely that even if music were not granted copyright protection innovation would still occur, especially since live performances cannot be easily substituted for by live recordings. Motion pictures on the other hand involve high capitol costs and would be impacted by a lack of protection. Now if that impact would be bad or good is a topic for another discussion.



My favorite new tweak in the world of song-tweaking, is bands making covers of their "own" songs when they have disagreements with their record label (who actually owns the song) about royalties: