Against Intellectual Monopoly

Against Intellectual Monopoly is a relentless, pounding, take no prisoners attack on patent and copyright law.  It joins Lessig’s Free Culture and Heller’s The Gridlock Economy as an instant classic and a must-read on these issues. 

Many people argue that the patent system has gone wrong in recent years, Boldrin and Levine argue that the patent system was rotten from the start.  James Watt they say was a "scoundrel" who with his politically-connected partner Matthew Boulton used the patent system to crush their innovative opposition and delay the industrial revolution. 

During the period of Watt’s patents, the United Kingdom added about 750 horsepower of steam engines per year.  In the thirty years following Watt’s patents, additional horsepower was added at a rate of more than 4,000 per year.  Moreover, the fuel efficiency of steam engines changed little during the period of Watt’s patent; however between 1810 and 1835 it is estimated to have increased by a factor of five.

Will books be published without copyright?  Boldrin and Levine point out that the 9-11 Commission Report was profitably published by Norton despite being available free for download. Not to mention the fact that most of the great works of literature were published without copyright.  Boldrin and Levine are top-notch theorists but AIM is widely accessible and it succeeds best with its many historical discussions and contemporary anecdotes.

AIM does suffer in places from a lack of a lack of nuance and a surprising ability to ignore trade-offs.  Boldrin and Levine argue, for example, that among the reasons we don’t need patents are a) because ideas aren’t copied immediately, they take time to diffuse, b) first movers have significant advantages and c) trade secrecy is often a more effective "means of appropriating returns" than patents.

Quite right on all three counts but each of these reasons also explains why patents are less costly than one might at first imagine.  After all, what Boldrin and Levine are really saying is that intellectual monopoly would exist even without intellectual property law

A standard model used to explain why patents might be useful implicitly
assumes that ideas are transmitted instantly at zero cost.  Boldrin
and Levine smash the premise of this argument but the premise is sufficient for the conclusion not
necessary.  Indeed, once you acknowledge that the slow diffusion of ideas helps entrepreneurs to appropriate the returns to their innovations it becomes an open question of how slow is best?   When is the appropriability of returns strong and when is it weak?  Doesn’t it differ for different goods?  Shouldn’t intellectual property law recognize these differences?  It’s clear, for example, that ideas are diffusing more quickly than ever before.  On Boldrin and Levine’s argument, faster diffusion of ideas implies lower appropriability and thus a stronger argument for intellectual property law.  Needless to say Boldrin and Levine are too busy using
a "mallet to smash shiny myths" to make this argument.  (To be fair, they are more nuanced in
the conclusion.).

Similarly, Boldrin and Levine argue that the larger the market the less patent protection is needed, hence globalization implies less patent protection.  Again, quite right (see also my paper, Patent Theory versus Patent Law, on this point).  But you won’t see Boldrin and Levine drawing the corollary conclusion that more intellectual property rights are optimal the smaller the market, despite the fact that we have a very successful example where increased patent rights for smaller markets generated considerably more innovation, namely the Orphan Drug Act.

For economists, it’s also surprising how little marginal analysis you find in AIM.  For example, Boldrin and Levine ask, Did Rowling really need a billion dollars to write Harry Potter?  Surely, a few million would have been enough.  But that’s like saying that taxing lottery winnings won’t reduce the number of buyers because the winner will still get a huge return on her dollar of investment.

The bottom line is that that there is a Laffer curve for innovation – more appropriability increases innovation at first but innovation declines when appropriability extends too far. I agree with Boldrin and Levine that rent-seeking has put us on the wrong side of the Laffer curve for innovation.  We need to reduce intellectual monopoly with patent reform, less copyright protection, and a greater use of patent substitutes like prizes.  But unfortunately, when it comes to innovation there is no invisible hand theorem which moves us automatically to the top of the curve. 


Most of your points seem perfectly OK, but I have some doubts about the Harry Potter paragraph. Are you suggesting that a billion dollar 'jackpot' for writing a children's book DOES lead to significantly more attempts to write one, compared to say a 10 million dollar jackpot?

For me, and I suppose most people, the difference between 10 million and a billion is so esotheric that it would hardly matter in any consideration to spend time writing. And as far as I know, most writers see earning 'a living as a writer' already as the 'jackpot', with the chance of becoming rich hardly an important factor in their considerations.

(... I know, I know, it's below the belt, but I can't help it) where can I download a free copy of the Boldrin/Levine's book?

Coca-Cola never patented the formula for Coke because it would have to reveal the formula and the
patent would eventually expire. Coca-Cola has an intellectual monopoly without any legal
protection. Should Coke be forced to reveal its formula? Should it be busted up by antitrust?
Other firms make their own colas; does that mean the cola industry is competitive?

Should universities drop all bans on plagierism?

Charles Dickens despised the US for printing his stories in newspapers without any compensation.
Was his resentment unjustified?

The US trade deficit with China is much higher than it need be because China is refusing to pay
royalties for US music and movies. Should the US be indifferent? Will the world get less music
and movies as a result of this stealing?

Also, you might note that Judge Richard Posner has written quite a lot on this topic.

A free copy of the book is available at:

Ozornik: There's this thing called should try it sometime:

The Laffer curve is a powerful concept. But, it may not be linear. We could easily back the cat into the wrong bag. You need to keep asking "what do we need now?" Kind of a theory of constraints for policy.

One practical benefit of patents is that it promotes publication of ideas so that NON-competitors can use the idea to come up with ideas in their respective field.

But, I'm basically anti-IP, or at least right now I don't think our problem is not enough of it. For big companies already getting paid, who use IP to smash competition, I don't think the free-rider theory applies.

I've also come to the conclusion that ideas are a dime a dozen and it's the execution that is hard. Having this bureaucracy around the idea makes the hurdle for execution all the higher. Also, there's a lot of luck with a good idea. The best technical solution often doesn't succeed in the market. Noone knows something will be good until it is tried. What we need is the best system that matches outstanding idea generators to excellent executors. I bet folks could get paid for their ideas as well or better through something like a prediction market or prizes.

If I felt that the little guy could use the system to get what's right, I might be persuaded otherwise, but I've been told by one little guy how it's often not even worth the expense and effort to try to get a patent. The big guys can tie you up while the lawyers ravage you through your hip pocket.

As usual, the opposite has evolved from what the system is sold to address.

Should've caught that - it was published in November 2006, IER. Here's the abstract: In their 2003 Lawrence R. Klein Lecture, Michele Boldrin and David Levine argue that intellectual property rights may be damaging to social welfare. As empirical evidence for their theory they offer James Watt's steam engine patent, claiming that it delayed the Industrial Revolution by as much as two decades. We show that this claim, as well as the more general claim that Watt's story supports Boldrin and Levine's theory, rests upon a distorted summary of the historical record.

Excellent post, Alex, most discussion of IP gets too dogmatic and yours is an excellent departure from the trend. Based on your review, it sounds like the argument is, "Don't use laws to slow the spread of (instantiation of) ideas, just sheepishly rely on physical constraints to do the job, and when they can't, too bad!"

For discussion: if ideas (specifically, the right to instantiate an idea) can't be owned, that means the "right" price for the market to place on ideas is zero. Yet in reality, we know people would pay a lot of money for certain ideas to simply exist, let alone for it to be instantiated in such a way that they can use it. How is it possible to reconcile promotion of the Mises/Hayek "Economic Calculation Argument" yet oppose IP? Inquiring minds want to know.

IMO if patent and copyright are not clearly but pluses for society we should drop them. Why bother if the advantage is debatable? Also Court costs should be a consideration. Also it seems that countries only enforce patent and copyright when it seems in their favor to do so.

Re: Watts, Boldrin and Levine were not the first economists to cite his patents as a barrier to the development of the steam engine.
Second, Selgin and Turner corrected a few factual errors, but didn't refute their charge that his patents were a barrier to competition in its development.
Re: the first mover advantage, it doesn't guarantee profits, market share, and business success. Remember the sock puppet, who now lives only on eBay? I'll bet does.
Remember WebVan? I'll bet FreshDirect does.
So a first mover advantage needs good execution in order to succeed. Boldrin and Levine don't deny this, nor do they claim that a first mover advantage guarantees a winning business. It also helps to sell complementary services, which they point out.
A first mover advantage is just that--an advantage. It's neither a necessary nor a sufficient condition to grow.

Ozornik stole my idea. i will believe then when they give away their book

Karl - they are giving away their book. I just downloaded it for free from Levine's web site, onto my Sony Reader. And I noticed that in their preface they acknowledge the corrections of Selgin and Turner, which did not, obviously, however, change the main point of their story about Watt.

"they are giving away their book. I just downloaded it for free from Levine's web site"
Not to split hairs, but does it say that you are free to use it as you will? If they are truly giving it away, then you could reproduce it and sell it yourself. Something tells me that Cambridge Press would think otherwise if you started selling copies of it for $10.

@liberty: Gee, thanks for explaining basic terminology, I *totally* must have been unaware of all that. Plus, I *totally* must not support an independent invention defense to copyright infringement, right?

The argument isn't that ideas have no value, or that they should not be protected with property rights -- it is that government shouldn't protect them with a monopoly privilege.

Right, and the argument isn't that we should cut down this forest, isn't that we should cut down all these trees over here.

Same. Damn. Thing.

Ownership of something *means* having a legal monopoly ("privilege") on it. If something doesn't have such a legal monopoly on it, it has no market value, as no one can actually buy or sell it, just seize in a war of might makes right.

So if there is no IP, that is a price ceiling on the market value of (the right to instantiate) ideas, of zero. Yet we know that for entrepreneurs to correctly decide on alternative uses of their property, they must have a higher signal than this for some ideas, such as a cancer cure.

(Btw, IP proponents DO NOT claim ownership of "ideas" but rather, "the right to instantiate specific ideas". Once you phrase it that way, you know, correctly, you lose all the fun in the non-rivalrousness charade you just threw at me.)

Now, try again, and see if you can do better than a word-switchout for your response.


Btw, I forgot: the reason it's profitable for them to sell physical copies of the 9/11 report is because they didn't have to fund its original writing. (I can't believe anyone ever actually advanced that as an argument.)

Modern patents are so busted that cyberspace is being stolen by lawyers, just like so much already-settled frontier land was as well. It's helping the lawyers alot more than the people, of course.

I think patents are OK, just WAY too long. They should be no longer than product cycles, because after that, they by definition slow product innovation, right? They should default to 2 years and be lengthened on a per-industry basis for industries with long product cycles (like drugs'long FDA-dictated cycles). Their period should also shorten 10% eery decade, to reflect shortening product cycles in general.

liberalarts, if you go to their website you can find on the left a copyright notice. You can download their book and sell it, no problem. You can even use copyright yourself with the book, what Cambrigde Press probably has done also. So I won't make a copy of the offline version and sell that, if I were you. The one think you cannot do is saying that you have written the book, instead of Boldrin and Levine. That is plagiarism and Boldrin and Levine are against it (but probably won't sue you).

Michael: How would an "independent invention defense" fail to prevent such a problem? That is, why would the problem arise if you could infring any patent on which you could document how you came up with it independently? It wouldn't be much of a burden for amateurs (beyond the unavoidable legal costs) because they could just send copies of their notes to a cheap third party storage company.

Person: if I can get out of the patent system by just documenting my efforts, and claiming that I never read patents, why can't a business do the same? Why wouldn't they just insist all their software developers avoid reading patents (which most do anyway) and then ignore the entire patent system?

The "independent invention defense" seems equivalent to just throwing out the idea of software patents. Or equivalent to eliminating most patents on "obviousness" criteria. I'm sure patents involving lots of hard math would still be granted, and still purchased by companies. Those are not the sorts of things you are going to just stumble over, or redevelop to avoid paying a royalty.

My problem though is more general. I take the law seriously, and I want to obey it. How can I invent something and know if I can use my own idea without lots of work and legal uncertainty?

With copyright, I have the information I need -- I know I'm the author of these words and haven't copied anyone. With patents, there is a large (and growing) body of information I would need to have, and don't have, in order to obey the law. And some of it is so opaque I don't think I could ever come to any firm conclusion.


"they are valued for different reasons"

How so? I understand that a legal right is a known factor for calculations; but it may not have additional benefit if the market-based right is secure enough. A legal right has additional problems too - such as the cost of enforcement. Many de jure rights in practice are less useful than de facto rights - it depends.

In addition, this de jure right will be imposed somewhat more uniformly - in hopes of providing additional security where the market-based right comes up short - and this will impose costs where it should not in fact exist. Like all government interventions to correct market failures, one must weigh the supposed benefits of correction against the potential costs - of government failure.

So, I just wonder whether you can say for sure that the legal right to monopoly over the implementation of the idea is in fact valued differently and that it is necessarily is a net benefit for quantity and quality of inventiveness. I don't know the answer - I am just posing questions.

I don't claim to have the answer to that.

"Then don't pretend to.""

I don't see that I have.

liberty wrote:
John Kay is trying to get around it, but it involves central planning

Yeah, really, I'd be just as happy if patents were abandoned, because they're SO open to abuse. But I see it as being in practice easier to drum up a political coalition for major patent reform. You don't have to tell congresscritters voting for the bill to give up all hope of drug industry money, for example.

@liberty: I can't even figure out what kind of system (or set of systems, or metasystem, of distinction, or metadistinction) you're holding up as an alternative now. You just distinguished a legal right (that I specifically allowed to be some non-state legal system) from a market right. What is a market right *but* a legal right? It seems you're doing the forest/trees thing again. When you start making sense, we can start making progress in our discussion.

@Michael: that they would run their development as now, document things to the legal requirement, and then just claim independent development if they were sued. So what's left of the patent system if no one can claim against you?

People *can* claim against you -- if you didn't independently develop the patented invention!

As for the rest, I absolutely agree that the current system allows obscurantism. I think it could be fixed by invalidating any patent for which there exists a simpler way of making the same claims.

@Andrew: extend to me a tiny bit of credit. Yes, the market actually *wanting* an intellectual work is necessary for the IP rights to it to have value, *in addition* to there being a legal system that actually enforces the rights. That should have been clear from context, or you're asking me to make every single background assumption explicit every single time.

As I see it, entrepreneurs have to calculate the value of the market competitive factors already. In the absence of IP, the market competition research might just be more important. If the legal value is zero, then that's an easy calculation. In fact, much easier than guessing which way the courts will rule next year or whenever.

And I'm sure you're consistent in bemoaning the extensive economic losses due to entrepreneurs wondering about *property* law rulings in the future, right?

As for the practical aspect, I'd suspect there may be less innovation in easy to copy technologies, but if they are easy to copy, then the argument is that the anti-commons is holding up easy innovations through prior restraint.

Easy to copy is NOT the same as easy to invent! Easy to invent shouldn't have IP; "easy to copy" includes medicines that take lots of time and money to develop. Do away with IP? Then you get to see libertarians become socialists as they list all the alternative, non-market ways the medicine would exist, which *obviously* are just as good. Oops! That argument applies to *all* goods. So, can we expect charitable groups to make our hamburgers now?

Regarding Bill Stepp's comment, although John Turner and I didn't pretend to refute their general claims concerning the stifling effects of patents, we do claim that Boldrin and Levine only succeed in making the story of Watt's engine appear to support their thesis by offering a factually inaccurate version of the story. That Boldrin and Levine also regard Watt's story as one that fits their theory especially well in turn supplies reason for doubting the theory's general validity.

It is enlightening !

I didn't see it mentioned here so. . .

The primary problem with the 9-11 Commission Report example is that Norton did not pay any of the costs associated with producing the work. The government paid for all of the research, the staff, the authoring, the copyediting, etc.

Had Norton born the costs, my guess is that it could never have made enough money from publishing the report.

A second problem with this example is that 2004 is not the same as 2009 and beyond.

I would bet that not too many people knew of the availability of the free PDF version of the Report.

Of those that did know, many did not want to read such a large book on a computer screen.

Of those who were comfortable reading so much on a screen, many would not want to be tied to a computer during their reading time and so would want a printed version which would allow them to be more mobile.

And then there are the subset of people who want to annotate the book, an option the PDF version did not offer (without specialized software one has to purchase).

All in all, these factors would decrease the demand for the free version allowing the paid-for edition from Norton to survive.

Why is this a problem and not a boon to the defense of a copyright-less Utopia? Because technology and attitudes are changing.

  1. Distribution of electronic documents is becoming more and more widespread.

  2. Time from release to dissemination is decreasing rapidly.

  3. Gaining access to documents from just about anywhere on the planet is becoming easier and easier as Internet access expands.

  4. Computer screens are getting better in terms of clarity and ease of reading (think Kindle).

  5. Also, devices are becoming more mobile, allowing people to disengage from their computer and still access electronic books (again, think Kindle, netbooks, and similar devices).

  6. Finally, more people are becoming used to the idea of reading from a screen and this will likely only increase as new generations and new technology make it easier and more familiar to do so.
Put it all together and the Norton 9-11 Commission Report example becomes hopelessly unique; it cannot be expanded into other areas and into the future as a defense of getting rid of copyright laws.

In the future, if a book is freely available online, the opportunity for profits will dwindle substantially compared to what is was in 2004 when Norton made its deal.

If you want to make the argument that getting rid of copyright laws is a good thing, go to town. But the Norton 9-11 Commission Report example is not a valid example of the point I think you (defenders of this view) are trying to make.

Comments for this post are closed