As noted on Marginal Revolution, in Launching the Innovation Renaissance, erstwhile quasi-Austro-libertarian fellow traveler Alex Tabarrok has a new book out in the intriguing TED Books imprint, entitled Launching the Innovation Renaissance (Amzon link, B&N for Nook, also iTunes). The description of the book says:
How can we increase innovation? I look at patents, prizes, education, immigration, regulation, trade and other levers of innovation policy.
Tabarrok is presented as some radical or maverick, bravely challenging the modern horror of statism and patent. But he is not really against patents. In the book he says:
Patents, innovation prizes, patent buyouts and advance market all have their place. The key is to match problems to institutions.
So patents “have their place.” The patent system should be reformed, but it has its place! Of course patent reform is both unrealistic, and not a solution, any more than tax reform is needed.1 The only real tax reform is to lower the rates, not to shuffle things around and move from one type of tax to another. Likewise, the most meaningful IP reform, short of abolition, is to reduce the length of the term: patents, from 17, down to a shorter amount like 5 years; copyright, from over 100 years, to, say, 10 or 20. (See my post How to Improve Patent, Copyright, and Trademark Law.)
As for the “prizes,” in his new ebook he highlights private prizes like the X-Prize but downplays the fact that he thinks taxpayers should fund these prizes. But this is the idea. As I have noted previously,2 Tabarrok is in favor of a taxpayer-funded “medical innovation prize fund”–starting at “$80 billion per year, and increas[ing] with the growth in GDP“. Similar proposals include those by faux free marketeers Joseph Stiglitz and Forbes.com. (Update: I’ve read more of the book now; he doesn’t downplay the taxpayer-funded aspect of the prize system he (and socialist Bernie Sanders) advocates. He is explicit about it in the book.)
Of course, medical innovations are only a small slice of the space of technologies allegedly promoted by patent law (there are electronics, lasers, chemicals, data processing, pharmaceuticals, and so on; there are over 400 classes in the PTO’s classification system, and each class is divided into numerous subclasses). So if you extend this tax funded innovation prize idea, and replace all patents for all technology areas with tax funded prizes, you’d have to advocate $2 trillion to $20 trillion a year in taxes to stimulate the “right” amount of innovation. Or maybe more. Hurrah for “free market” “solutions” to our “problems.” What the hell, let’s be “bold” and make it $100 trillion of tax funded innovation prizes per year to create a utopia on earth by 2013! Or maybe a quadrillion dollars!
Sorry, did I say “replace”? As patents have their “place,” these prizes would not even replace the patent system, but supplement it. Injury upon injury! In this, I am reminded if calls for “replacing” the current income tax with a VAT or national sales tax. Of course, in practice this amounts to a call for adding a new sales tax on top of the current income tax, since the state will never give up the latter.3 Likewise, Tabarrok’s call for a taxpayer funded prize system would not result in this replacing the patent system, but being added on top of it, making things even worse.
Tabarrok has done good work before; see Tabarrok on Ideas and Prosperity. But this centralized, statist approach would merely make the current system even worse.
N.B.: My Copy This Book: The Case for for Abolishing Intellectual Property is forthcoming next year in a similar compact ebook format from Laissez Faire Books. It’s highly unlikely my book will advocate taxpayer funded innovation prizes.
Update: Tabarrok replies in Tabarrok the Statist, where he says that I call him “a liberty-hating statist!” and that I am “upset that I did not come out as a patent abolitionist.” I did not call him a liberty-hating statist, nor do I think he is; I don’t know much about him other than his confused views on IP and innovation, and the good work he has done before, which I noted above. Rather, the title of my blog post intimates that the policy measures he advocates to increase innovation are statist: namely, the patent system (which he wants to restrict but not eliminate, as it “has its place”) and billions of dollars of taxpayer funded “innovation prizes.” Nor am I “upset” that he doesn’t want to abolish patents; rather, I disagree with him and believe his “solution” is not a just or effective way to promote innovation. (I’m not too happy that he wants the state to steal money from me and let a panel of “experts” award it to “deserving” innovators.) The best way to promote innovation is to reduce the state’s size and scope, so that the free market flourishes and more wealth is available for R&D.
I have read more of his book now. It has some good ideas (allow more high-skilled immigrants in), and some fairly conventional and tepid ones (improve how teachers are paid; implement school choice). Some of his criticisms of the existing patent system are good, but, as noted above, his proposed policies are confused and unlibertarian. In his post, Tabarrok notes that Bryan Caplan says Tabarrok’s book “is a gem that pushed him to extreme patent skepticism.” In his comments, Caplan says: “Tabarrok’s patent litigation diagram alone is worth the price of admission; it genuinely pushed me over the edge to extreme patent skepticism.” But the patent litigation diagram, on “The Smartphone Patent Thicket,” is taken from a post by Mike Masnick on Techdirt (I have it in this post: Intellectual Property Charts, Graphs, Graphics, Diagrams).
In his chapter on the trademarked and patented Harvard “Oncomouse,” Tabarrok writes:
… patents increase the benefits of innovating but, especially in fields of cumulative innovation, patents also increase the cost of innovating. Since patents increase the costs of innovating, restricting patents does not necessarily harm innovators. When patents are restricted, firms lose some of their monopoly rights but they gain the right to use the innovations of others. The result is greater total innovation. [Kindle Locations 234-237]
It is true that patents help some companies, while they hurt others. It is also true that if you weaken or restrict patent rights, some firms lose some of the monopoly advantages they had before, while they benefit from being more free to use others’ innovations. But I do not see how it follows from this that if you weaken patents you get greater innovation in total. (I do believe it’s true, but it doesn’t follow from what Tabarrok says.) In any case, if Tabarrok thinks that restricting patent rights leads to greater innovation overall, why does he not want to abolish patents altogether? Why do patents “have their place”?
Innovators need time to recoup their sunk costs, but why should every useful, non-obvious and novel idea be granted a 20-year patent? Maximizing innovation requires treating different industries differently. The idea for one-click shopping does not have the same sunk costs of research and development as a new pharmaceutical, and the former does not need and should not be given the same monopoly rights as the latter. [Kindle Locations 241-243]
It is good that he openly admits that the patent rights he is in favor of are monopoly rights. Not all advocates of IP do. (See Are Patents “Monopolies”?)
Perhaps the most outrageous patent practice is that patent claims can be changed after a patent is filed to retroactively cover a competitor’s invention! Rambus, for example, filed a patent on dynamic random access memory in 1990, and two years later it joined an organization creating a memory-chip standard. By using a patent continuance, Rambus was able to effectively change its claims to cover the new standard even though the standard was designed after Rambus’ invention (and before Rambus’ patent was issued). [Kindle Locations 275-279]
I am as big an opponent of patents as anyone, but I don’t really get this criticism. You cannot add new matter to a patent application after it’s filed. But you can amend the claims so long as they are supported by your detailed description (specification). And amending claims can be done even if you don’t file a “continuance” (we patent attorneys usually refer to this as a patent continuation, not a patent continuance). It’s just that if you change the claims too much during prosecution you can get a final rejection; the continuation lets you re-file it and keep going. In essence, it’s simply another fee you pay to the PTO to get them to examine a different set of claims. This kind of practice is par for the course. If you have a patent system based on claims, then so long as the system is one in which your application is examined and then granted, resulting in an issued patent with a presumption of validity (as opposed to a petty patent or utility model system in which there is no substantive examination and thus no presumption of validity; it has to be determined later, during litigation), then there has to be prosecution: the back and forth between the patent applicant and the PTO. And this means the applicant has to be free to modify the claims. And you cannot limit the number of iterations, otherwise some applications would be arbitrarily ended because of a stubborn examiner. So you have to allow it to go on and on. But the PTO does not want to do it for years for only one fee, so you will have a system where additional fees have to be paid every few iterations–hence, the continuation system. And given all this, the applicant can of course gradually tailor the claims over time as he sees what products competitors have. This is all part and parcel of our type of patent system. It is not outrageous at all; it is a natural consequence of the patent system that Tabarrok is in favor of.
One problem that used to arise because of this use of continuations was the “submarine patent” problem. As I explained in Reducing the Cost of IP Law:
Until 1999, patent applications remained secret until they issued. Thus arose the problem of “submarine patents“: patents could remain pending in secret for decades — after industry had independently invented and widely adopted the technology — and then “emerge” like a submarine and extract heavy royalties from many companies. The 1995 amendments to patent law changed the patent term from 17 years from date issuance, to 20 years from the date of filing, to reduce this problem. And starting in 1999, US patent applications are now made public 18 months after filing, unless the applicant requests nonpublication and promises not to file the patent internationally.
So, the submarine patent problem is mostly gone now (though not entirely — see Submarine Patents Not Dead Yet).
In discussing the America Invents Act, Tabarrok writes:
The America Invents Act (H.R. 1249), although labeled as patent reform, did very little to improve the patent system. The act includes enhancements to prior-use rights but this is limited to business-method patents for the finance industry, a clear special privilege for this industry alone. [Kindle locations 309-311]
This is not correct. As I explain in The American Invents Act and Patent Reform: The Good, the Meh, and the Ugly (see the entire lecture at My Webinar on the America Invents Act), before 1998, there was no prior user defense at all. In 1998 a narrow exception was added to give “prior users” of business methods a defense against those who later patent the same method. Subsequent caselaw weakened business method patents so that this defense was even more rarely needed (see the Bilski case in 2008).
Contrary to Tabarrok’s description of the AIA, it actually adopted a fairly broad prior user right, one not restricted to business methods. I think Tabarrok here is confusing the previous prior-user defense, which was limited to business methods but which is now broadened by the AIA, with another provision of the AIA, to-wit: tax strategies and some financial methods are declared unpatentable by the AIA. This is a (minor) improvement, albeit arbitrary: this exception appears to be the result of special-interest lobbying by the banks against Claudio Ballard and his company Data Treasury who used patents to extract royalties from banks who “violated” its check image capture and storage technology. (For more on this see my post Banksters versus Patent Monopolists.)
Tabarrok goes on:
The bill has a number of other provisions, such as stipulating “first to file” rather than “first to invent,” but although much debated, this is a trivial adjustment that will affect very few firms. [Kindle Locations 311-313]
This is correct. Tabarrok is one of the few to recognize this. Patent proponents and practitioners have tended to refer to this as some major change.
He goes on:
In many areas, the bill enhances patents by such means as including provisions for treble damages, which will increase litigation costs. Perhaps the most beneficial aspect of the act is that patent fees will flow to the patent office, and such fees could be used to improve the speed and quality of examinations. [Kindle Locations 313-315]
I am not sure what he means about treble damages. This has been a part of the law for a long time. In 2007 the Court of Appeals for the Federal Circuit (CAFC) — the sole appellate court for patent cases since its creation in 1982 — changed the standard for “willful infringement” in the Seagate case, making it harder to obtain enhanced (treble) damages. I am not aware of any provisions of the AIA that have to do with willful infringement or enhanced damages.
As for Tabarrok’s latter comment about patent fees, he seems to think that one problem with the patent system is the low quality of patent examinations, which can be improved by increasing PTO funding. But complaints about PTO “fee diversion” are very confused, as I noted in Patent Fee Diversion and Patent Reform Whining, where I wrote:
It is not true that everybody agrees fee diversion is a bad thing. The money collected by the PTO is somewhat arbitrary given the monopoly power (to grant monopoly patent grants) given to it by the state, and given that the PTO is basically “owned” by fedgov. Just as a parent company may do what it likes with a subsidiary, so here. It is not clear at all that all of the “profits” of the subsidiary-PTO are necessary to fund its operations. Suppose Congress ordered it to quadruple patent filing fees tomorrow. This may well result in a huge windfall. It is not immediately obvious that all the surplus profit would need to be diverted to the PTO for spending on operations. The idea that all of the fees the PTO generates (using the state monopoly granted to it) need to be kept by it to fund operations presupposes that it will never make a profit.
The opposition to fee diversion also rests on other controversial assumptions–namely, that it is good that the PTO be well-staffed and funded; and that the PTO itself is a good thing. For someone who believes patents are harmful and a plague that should be ended (and believe me there are many companies besieged by patent assaults who think this), they might prefer the PTO to be starved of funds to hasten its demise and/or the reduction in the perception of patent quality/legitimacy.
Several places in the book Tabarrok suggests that the taxpayer-funded prize system could replace the patent system–e.g., “Prizes could take on a much bigger role, however, if patents were converted to prizes.” [Kindle Location 345] Yet elsewhere he says patents have “their place” and in his blog post he makes it clear that he is not a patent abolitionist. But if the prize system replaced the patent system, that would abolish it. Elsewhere he suggests the prize system would only supplement, not replace, patents: “Prizes can also be used to supplement patents where patents are least effective.” [Kindle Location 381] So I’m confused as to where he stands.
Regarding patents for pharmaceuticals, Tabarrok writes:
GlaxoSmithKline sells the AIDS drug Combivir in the United States for about $15 to $20 per pill. Without patent protection the price would fall to the cost of production, about a dollar per pill.40 Activists outraged by high prices for life-saving drugs demand that pharmaceutical patents be abrogated or price controls be imposed. In the case of pharmaceuticals, we have seen that innovation-to-imitation costs are high, so abrogating the patent or imposing price controls would reduce innovation. [Kindle Locations 346-350]
But as Boldrin and Levine show in ch. 9 of Against Intellectual Monopoly–which Tabarrok cites elsewhere in his book–it is simply not true that abrogating patents would reduce innovation in the pharmaceutical industry.
Regarding the prize idea, he writes:
Is it a good idea to throw a medical innovation fund into the budget battle with every other claim on public monies? How will a medical innovation fund fare when the country is at war? The prize fund could be a tempting target for budget cutters who wish to spend on other goals (“just temporarily,” of course). Pharmaceutical profits already evoke the ire of voters. What will happen when these profits flow from directly from taxpayers? If voters are myopic or do not understand the long-run benefits of pharmaceutical innovation a prize fund could be a disaster.43 [Kindle Locations 365-369]
43. On voter myopia and other irrationalities see Bryan Caplan, The Myth of the Rational Voter, (Princeton: Princeton University Press, 2007). [Kindle Locations 939-940]
Actually this is the one good thing about replacing patents with a prize fund: the theft that patents now result in, in a disguised way, would be out in the open, and could possibly be reduced over time, as voters begin to resent the taxes or politicians reduce the prizes due to budget pressures. The last line, however–calling voters who are being robbed to the tune of at least $80 billion a year “myopic” and too simpleminded to “understand” whey they need to be robbed–reeks of the condescending arrogance of the planning class. The dumb citizens should shut up and pay the taxes that the experts declare are “necessary.” The sheeple are good only for making money and funding the benevolent state–not at thinking, damnit!
And notice the cockamamie scheme proposed by Michael Kremer that Tabarrok endorses:
The Medical Innovation Prize Fund, as a mandatory replacement for patents, has two problems: It’s a) difficult to estimate the true value of a patent and b) difficult to avoid politicization of the reward process. The economist Michael Kremer has made a clever proposal that avoids both of these problems.44 Kremer suggests that patents be auctioned, much like electromagnetic spectrum bands or timber licenses are auctioned today. In an open auction with plenty of bidders, the winning bid will be a good estimate of the true value of the patent. In Kremer’s proposal, after holding the auction the government will then roll, say, a 10-sided die. Nine times out of 10 the patent would not be sold to the high bidder but to the government at the auction price plus a markup. The markup can be used to increase the incentive to innovate and to encourage patent holders to offer their patents for sale voluntarily. One time out of 10, the patent is sold to the high bidder at the auction price.
Kremer’s patent-buyout process has two virtues. First, nine times out of 10 the government buys the patent and throws it open to the public, thereby opening the market to competition and lowering prices. Second, the auction price also reveals the best estimate—and a nonpoliticized estimate — of the value of the patent, which combined with the markup, means that patent buyouts increase the incentive to innovate. The one time out of 10 the patent is sold to the higher bidder maintains the incentive to bid carefully. [Kindle Locations 369-380]
“It’s difficult to estimate the true value of a patent”–you don’t say! And it’s “difficult to avoid politicization of the reward process”–how shocking! Kremer’s proposal is not “clever”–it’s ridiculous social engineering. It results in keeping an arbitrary one tenth of patents in force. And just imagine companies bidding on a patent that they have only a one-tenth chance of getting, even if they have the highest bid! Is this supposed to seriously be a pro-liberty, pro-property, pro-innovation, or pro-free market approach?