Comments for Do patent rights facilitate or impede cumulative innovation?

Like Newton said, every major innovation is inspired by several others. That’s what we called cumulative innovations. Whether sequential or complementary (based on an artice written by Paul Belleflamme, “A primer on cumulative innovation”), all innovations are based on old.

We all know that the patent system is an evil necessary, as many authors like to call it. It allows first of all to avoid the well-known tragedy of the commons, which refer to the fact that a ressource can be overused if she’s not well protected. It also give the good incentives to people who wants to create some new technologies by giving them the insurance of a temporary monopoly of using. But let’s return to the problem of cumulative innovations, is patent system facilitate or impede this phenomenon ?

The patent system give to the inventor the right to produce (or not) his product, but it also give him the right to deliver licences. As the owner of the patent, he has the power to choose with who he is gonna work and who can be authorised to use his innovation. It’s a way to reduce the diffusion of the innovation, especially if the futur inventors have to wait until the end of the protection. It also have a cost, like a transaction cost, mostly if the inventor need several licences to start his production. And, obviously, sometimes, the addition of this costs is higher than the expected profit, an other phenomenon that discouraged innovation. There’s solutions to that problems, like pools of patent and cross-licencing, but it’s fixing the problem after its appearance, so why not fix it at its source ?

Galesso and Schankerman tried to determinate empirically the real impact of the patent system on the cumulative innovations. They managed to fix three main problems that comes with an empirical experience and their results let’s to think that the invalidation of a patent would increase the rate of cumulative innovations.

So, to conclude, I think we can say that patent system is imperfect and in this case, is impeding cumulative innovations that are so important to economy. We can also observe this pattern in other situations, like in the medical industry, where several companies own patent on drugs or on their composition. This informations could be used by others and lead to develop some treatments faster and probably save lifes, but the patent system give the right to this companies to keep their researchs secret and protected during a time, allowing them to sell themselves, and only them, the treatment.

We can also quote some situations about land ownership. Heller talked about a situation (in his article “The tragedy of anti-communs”, 1998) after the fall of communism, where several cities had empty and unexploited premises. The traders on the spot would have wanted to used them but because of the system of propriety, that protect the owner, it was hard for them to get the right to use this unexploited places. Although all the owners of the spaces were losing money with the empty stores, while the commercial areas were in great demand, their competing interests prevented the effective use of the space.

In conclusion, the patent system is uneffective about cumulative innovations, but is that due to the system himself or to the bad use of this one ? Some authors advanced the theory that this problems actually came from the large number of patent on the market, a hypothesis which would benefit from being thorough.

Legal protection of cumulative innovation is a major problem in the field of intellectual property because a large number of inventions have multiple applications. Most industries today are part of a process of incremental progress. While current inventions allow the emergence of new products, they are at the same time the bedrock of future innovations. The patent regulation then plays a crucial role since it is for the legislator to balance the interests of the “pionnier” innovator with the interests of innovators of subsequent generations in order to provide them with the best possible incentives. If patents are very extensive, pioneers are encouraged to develop their innovations, but this to the detriment of innovators of subsequent generations. If patents are restricted, the situation is reversed. How to encourage one without disincentive the other? What kind of patent regulation need to be in place to provide the best incentives for innovation? some authors like Edmund W. Kitch and Chang H.F believe that legislation that favors a broad patent conferring a monopoly for the pionnier innovator will have a beneficial effect on incentives to innovate. On the other hand, authors such as Robert P. Merges and Richard R. Nelson take the opposite position and explain that such legislation has detrimental effects on the incentives to innovate. They evoke arguments that complement those contained in the article.

Edmund W. Kitch was one of the first authors to propose, in 1977, a complete theory of the role of patents in cumulative innovations. It relies on the “tragedy of commons” to explain that while information is not subject to scarcity, this is not the case for resources available to manage this information. He emphasizes that giving the pioneer innovator an exclusive right to exploit his technology allows him to better coordinate his future research and development efforts. In this case, no one can invest in the development of his technology without obtaining permission. Competitors will be encouraged to share information instead of getting tired of working individually in a discreet manner without taking into account the efforts of others. This coordination phenomenon thus makes it possible to “optimize the flow of innovations over time” as stated by Bruno Deffains.

Chang H.F extends the work initiated by Kitch by presenting another argument in the same direction. He observes the existence of a distortion between the “social value” of the pioneer’s innovation and its “intrinsic value”. For him, an invention may have a low value for itself but an important social value if it opens the door to future progress. As a result, this author believes that the pioneer innovator is “legitimate” to obtain a part of the “social surplus” of innovation generated by the progress made thereafter. The solution remains efficient even if the costs generated exceed the intrinsic value. This distribution of surplus is, for Chang, both the “just reward” for the efforts of the pioneer but also a determining factor to encourage the latter to develop his innovation and distribute licenses.

If these arguments seem to be strong from a theoretical point of view, they do not take into account the fact that innovation is a dynamic process resulting from what Bruno Deffains calls “the existence of a competitive structure at different stages of innovation development “. To put it simply, the pioneer innovator is not always in the best position to develop his invention, or does not necessarily have the technical skills. It is on this basis that the detractors of the pioneer’s exclusive right to exploit his innovation come into play. from the arguments presented in the article against exclusivity, such as the risk of “Hold-up”, the “licensing difficulties” or the “tax right” conferred on the patent holder, I would like to present three others that come to deepen or complete them.

Robert P. Merges and Richard R. Nelson first put forward, in 1994, an argument based on “cumulative systems technologies”. According to them, these technologies have two particularities. The first is that they evolve in a sequential process away from the initial innovation. The second is that they consist of different elements that can be the subject of multiple applications. Thus, conferring an exclusive right to the pioneer on its innovation gives him the right to collect very significant profits on innovations introduced by competitors and sometimes far removed from its initial innovation. Therefore, if the pioneer has the exclusive right to exploit a component that has been patented, he is able to paralyze the entire industry by blocking the appearance of technologies on the market using this component.

Secondly, our two authors explain that this exclusivity situation is perhaps even more problematic when it does not take into account how the product that is the subject of a patent has been manufactured. Competitors may use a patented process without using it under the same conditions. For example, if a company is granted an exclusive right of exploitation over all products using the same chemical process as the one he has just patented, then this patent absolutely prevents all competitors from using the basic principle of the process under completely different conditions. These first two arguments show us that a risk of market immobilism appears because each company tries to block others.

Finally, Merges and Nelson pursue by uncovering a new “innovation brake” stemming from the exclusivity conferred on an innovation using a principle based on public scientific knowledge. When researchers publish a scientific discovery, companies fight to patent it first if an industrial application is possible. If Company A uses the new process to release a technical innovation and patent, then it has the power to prevent a company B to use the scientific breakthrough to innovate in turn. Company A can attack for counterfeiting all companies that use the same process by a phenomenon of “privatization of the use of public science”.

To conclude, if we saw in this comment that there were arguments both for and against the exclusivity conferred on the pioneer, we can draw the conclusion that there is no regulation able to optimally incite all the actors at the same time, and that the incentives of one causes more or less disincentives for others. While the extended patent did not prevent the appearance of a wide variety of products on the market, it is, in my opinion, still an obstacle to innovation. The role of the legislator is thus to find “the optimal balance” between the incentives to innovate for pioneers and for next innovators.

Sources :

Deffains Bruno. Progrès scientifique et analyse économique des droits de propriété intellectuelle. In: Revue d’économie industrielle, vol. 79, 1er trimestre 1997. L’économie industrielle de la science. pp. 95-118.

Santagata Walter. Propriété intellectuelle, biens culturels et connaissance non cumulative. In: Réseaux, volume 16, n°88-89, 1998. La propriété intellectuelle. pp. 65-75.

Journal of Economic Literature vol. XXVII (September 1989) pp.1067-1097

Journal of Law and Economics, Vol. 20, No. 2 (Oct., 1977), pp. 265-290

We have two different types of cumulative innovation.
First there are, what we call complementary innovations. These are the generation products, which are based on a mixture of the components from several early generation innovations (from multiple innovations to one)
For the other category we use the term sequential innovation. This means that we have first-generation products, which lead to second-generation innovations. If the first one doesn’t exist the second could not exit.

The question if patent rights facilitate or impede cumulative innovation is in a direct relation with the breadth of the patent.

Patens are likely to block cumulative innovation as shown by several studies.
When the breath of the patent is too expanded the second-generation innovations will possibly not take place, because the costs are too high for the innovators. But contrarily, if the protection isn’t sufficient the number of first-generation will probably decrease, which may decelerate economic growth.
We have three different solutions to face this problem:
-In the first case the breadth isn’t a problem because the first and the second-generation’s innovator is the same.
– Another solution is to regulate the market of licences, which the innovator sells on his patent. The patent system could create a system for the licence market and put its hand on the price.
-And at least we have the solution to give up totally (or in several domains). This would induce that every innovator could profit from the research of the others. But all the innovators would be forced to share their information.

https://www.ipdigit.eu/2014/10/a-primer-on-cumulative-innovations/

Common anticommons?
For Buchanan and Yoon (2000) “Anticommons is a useful metaphor for understanding how and why potential economic value may disappear into the “black hole” of resource underutilization”.
They argue that an anticommons problem occurs when several entities have the right to exclude.
Heller and Heisenberg (1998) present anticommons to explain why merchants bought kiosk on the street meanwhile buildings remain empty in the 90’ in Moscow. More generally, housing permits are anticommons because they require the approval of several entities who all have the authority to prevent you to build.
James Buchanan presents another example of anticommons. An entrepreneur wanted to invest in a seaside/hunting-preserve resort. His project was inhibited by the necessity of getting permits from several agencies. Even the environmental movement know the effect of interposing additional authorities with the right to exclude development of facilities. All these examples show the pervasiveness of anticommons in our economic interactions.
Does this have an impact on cumulative innovations? Do patent rights facilitate or impede cumulative innovation?
In the first instance, we defended that without patterns or copyright the intellectual works would not be produced. But, in the last decade, this rational has been strained. Indeed, we consider that the creator could appropriate other sources of profits with his intellectual work. The professor José Louis Ferreira tried to understand the various aspect of the problem. For him, patterns are balanced between efficiency and fairness. But knew complications are emerging with new technologies. Most of time, new inventions need the use of many previous development, each of them already patterned. The fragmentation of patent ownership and the asymmetric information on costs and benefits increases the costs of negotiation. This lead us to a patent system who is a handicap for innovation.
Sampat and Williams (2015) have worked on this subject for some time. They have analyzed data on successful and unsuccessful patent applications in the gene sector. Based on this data, they made a simple comparison of follow-on innovation across genes claimed in successful versus unsuccessful patent applications. After that, they used the leniency of the patent as an instrumental variable for whether the patent application was granted a patent or not. They found out that patent had not quantitively important effect on follow-on innovation.
To conclude, the patent rights have an impact on cumulative innovation. Anticommons affect negatively cumulative innovation, but otherwise we can not predict how the patent rights will affect the innovation.

Bibliography:
Buchanan, James, Yoon, Yong J.(1999). “Majoritarian Management of the Commons. Unpublished manuscript”.
Heller, & Michael, A.(1998). “The Tragedy of the Anticommons: Property in the Transition from Marx to Markets.” Harvard Law Review 111: 621-88.

Ferreira, J. (2015). How patent rights affect cumulative innovation. Retrieved from http://mappingignorance.org/2015/04/17/how-patent-rights-affect-cumulative-innovation/.
Sampat, B., & Williams, L.(2015). How do patent affect follow on innovation? Evidence from the human genome. Retrieved from http://economics.mit.edu/files/9778.

Throughout the years we have had to face a well-known phenomenon, the resources depletion due to the exhaustive use of common resources. However, besides the overuse of common resources, the underuse also forms a problem and is caused by the large number of patents. Could that be an indication that our current patent system is defective?

The underuse of resources, also known as “the tragedy of the anticommons”, is an issue that has become more obvious with the raise of new technological inventions that required previous patented developments. But, how has this problem emerged? When there are too many patent owners who exercise their right of exclusion, they inhibit the use of common resources. The high fragmentation of patent ownership is an obstacle for cumulative innovations because the innovators need to obtain various licenses which increase the costs of their new product. This often leads to the abandonment of the production plans for it. On the other hand, the high fragmentation is also a handicap for the patent holders. They lose their licensing fees because of it.
The tragedy of the anticommons is something we need to correct to avoid the underuse and thus the waste of common resources that could be crucial in the development of a new product. This issue questions the validity of the whole patent system, because it provokes the opposite effect of the patent system’s initial objective, that is to create incentives to push innovation.

The most common arguments used to point out the negative effects of the patent rights on cumulative innovations are the following. The first is the empirical study done by Galasso and Schankerman. In this study, they found out that the removal of patent protection increases the use of it in citations of subsequent innovations with 50%. The second argument is the fact that the increased number of patents resulted in holdup strategies. Finally, the last argument is the Myriad case where the Supreme Court decided to restrict the patent rights of the company, Myriad, to promote cumulative innovation in the genes sector. However, no generalisation of those negative effects has been proven through the analysis. We only know that the patent rights have a significant impact on cumulative innovation in fragmented and complex sectors, such as the IT sector, the software spaces and the biotechnology sector. Nevertheless, there are also counter-arguments such as a study on human genes where the effects of patent rights on subsequent innovation appeared to be very small.
In my opinion, the solution to avoid any ambiguity about whether or not the current patent system affects cumulative innovations is to redefine, in a more precise way, the framework of the patent rights and to establish targeted regulations on patent rights for specific sectors. By doing so, it will enable innovators in complex sectors, such as the biomedical sector, to base their innovations on previous ones. In those sectors, it is more important to innovate than to respect the rights of the patent holders.

Apart from that, I think that even if anticommons are a negative effect of the patent rights, when we look at all the advantages coming from them, it outweighs this little disadvantage.
I agree the patent rights should not freeze cumulative innovations, but that could be easily rectified with designing policies, institutions that facilitate more efficient licensing and targeted regulations.

To conclude, the anticommons issue is still very new and very difficult to identify. The worst we can do is to rush to hasty conclusions. We need to further analyse the problem and to find a mechanism that identifies the real impact of it on the different sectors before taking measures.

http://www.newyorker.com/magazine/2008/08/11/the-permission-problem
https://ideas.repec.org/p/wdi/papers/1997-40.html
http://mappingignorance.org/2015/04/17/how-patent-rights-affect-cumulative-innovation/
http://wealthofthecommons.org/essay/tragedy-anticommons
http://economics.mit.edu/files/9778
http://www.digitopoly.org/2014/07/17/do-patents-stifle-cumulative-innovation/
http://www.wpz-fgn.com/wp-content/uploads/1460556705_SZ6itWqcy9jo.pdf

I agree with the article that royalties may prevent new innovators from utilising the existing ideas which have already been patented. However, according to me this becomes an opportunity to unleash one’s creative potential in the industry which is essentially the idea behind protecting IP rights as well.
Take the audio-visual technology developments in the entertainment industry. If every new innovator decides to utilize the existing patents without any royalty then they will be tempted to carry out incremental innovations only. Another reason for this would be lack of reward at the end of the innovation (since every subsequent innovation will not be allowed to charge a royalty fee). Mark Zuckerberg would never have invented real time touch sensing videos had he decided to invest in incremental changes only. Thus, if royalty fee is a deterrent, avoiding it is also an incentive to innovate out of the box. Overall, such creativity would benefit the industry at large.

Assuming, it becomes indispensable to utilise an existing patent and a new player is forced to pay the royalties, then I suggest the following method to incorporate fairness in the system:
If a party utilises an existing patent during its research, it must gladly pay the licensing fee. Later, if it is able to achieve a patent of its own then it must be allowed to prove the utility of the prior patent in retrospect. The court of law may allow partial refund of the licensing fee if there is merit in the new innovator’s claim about the utility of the old patent. This will eliminate the stacking effect, and hence lead to lesser overall costs of the new product/ idea. Consequently some benefit will pass on to the customer as the innovators at each step could save costs.

In my point of view, I think that patent rights can do both (facilitate and impede cumulative innovation). Let me explain, as it is mentioned in the article “A primer on cumulative innovations” from Paul Belleflamme, it exists two types of cumulative innovation:
(1) “one to many” which are sequential innovations and (2)“many to one” which are complementary innovation. Those two types of cumulative innovations combine with patents’ issues, can lead to two different problems which are ‘hold-up’ for sequential innovations and ‘tragedy of Anticommons’ for complementary innovations.
In this view, patent rights can be seen as a brake for cumulative innovation because companies doing those types of innovations might not have the means to struggle those issues. Because if you are a small company, you will have less equipment to fight against those problems than if you are a bigger one.
Taking the case of those smaller firms, they would have less capital to invest for paying the licenses that give them the right to use the previous innovations. Moreover, the more patents they will have, the more they will have to pay to negotiate which make them less able to find new innovations and that give them less motivation to innovate.
I think that a good solution for that will be to use the patent pools but those are difficult to form and to implement.
However, the patent rights opportunity can be seen as an incentive to motivate firms to innovate because the more the firm will have patents, the more independent the firm will be because the company would not have to negotiate the transaction cost to gain licensees considering the fact that they will have their own. In that view, I believe that it will give them more incentives to find cumulative innovation.

To conclude, for me the debate on patent rights is not yet finished, I think that every firms or individuals should make a kind of list of which are the pros and the cons to file a patent and to decide whether it is better for them to use their own patent or to pay for licensee.

References
1. Belleflamme, P. (2016). Sequential innovations (slides). Not published, Louvain School of Management, Louvain-la-Neuve.
2. Belleflamme,P.(2014). A primer on cumulative innovations. IPdigit
3. http://avocats-publishing.com/LES-PATENT-POOLS

I want to contribute to this article by taking a theoretical approach on how patent rights influence cumulative innovations.

To find out what impact patents have on cumulative innovations it is necessary to have a look at the change of expected costs and expected profit between having and not having patents. If expected costs and expected profit don’t change, then patents would have no to little impact on cumulative innovations. However, if expected costs and expected profit change, which is closer to reality, patents have an impact on the incentives of innovators.

I’m going to discuss the impact of patents on cumulative innovations by differentiating between drastic and non-drastic innovations:
If the innovator expects his new innovation to be a drastic innovation he knows that by having sufficient low production costs with a high chance he can charge a monopoly price. This means he is very likely to make a profit even when paying fees to the owners of patents he used for his innovation. Patents for cumulative innovations in this area should have no negative effect on the incentive to innovate.

On the other hand, if the possible new innovation is expected to be no-drastic patenting has an effect on cumulative innovations. For this case, I separated innovations in two groups: innovations in fast developing industries and patents in slow developing industries. To be able to compare small and fast developing industries I do both analyses inside a timeframe with an equal length of 1.

Starting with cumulative innovations in slow developing industries, one can see that because of a slow development in the innovation flow sheet many “parent patents” in the considered timeframe are already expired. This leads to a small amount of active “parent patents” and therefore, to expected low costs for the innovator in comparison with innovations in fast developing industries. In my opinion, this expected cost doesn’t have a huge influence on the expected profit; hence the effect on the incentive to innovate is a slightly negative one.

To explain the higher costs in fast developing industries I’m looking, as in the previous paragraph, at the amount of granted patents inside this timeframe with the length 1. Because of fast development, the amount of innovations produced in this timeframe is huge, which for a new innovator means that there are many “parent patents” in the innovation flow sheet that are still valid. This implies that by having a great number of valid patents the expected costs also will be high. This in combination of a non-drastic innovation leads to situation where the innovator doesn’t know if the expected outcome will be a positive or negative. I think this will have a negative influence on the incentive to innovate. But, and this is a big one, patents can under certain assumptions have a positive influence on inventing. As mentioned before, in a fast developing industry there are many innovations made in a small time frame, therefore an innovator can think that there will be new inventions using his innovation after he got his invention patented. If he expects these future inventions to be made, patented and produced within the duration of his patent and assuming that the new inventors have to pay him fees which cover his innovation costs he is more likely to innovate even in a non-drastic, fast developing industry.

To conclude, I arrive to the finding that the effect of patents on cumulative innovations differs from sector to sector. This is similar to the finding made by Galasso and Schankerman (2013), however, my finding indicate that patents can accelerate the amount of innovations made in fast developing industries, in “sectors with complex technology”. This would mean that an abolition of patents wouldn’t increase but decrease the incentives to innovate.

It appears pretty clear to me that the theory of the anticommons is a serious matter that have direct effects on cumulative innovation. The initial objective of patent system is undoubtedly to encourage innovation in general, and cumulative innovation seems to be one of the weaknesses of the patent system. Throughout this comment, we will verify whether it is a common issue and I will explain why, in my opinion, patent system have opposite effects on cumulative innovation.

It is difficult to assess whether the tragedy of uncommons is global issue that affects all markets or just some specific markets. As it was mentioned in the article above, “the positive impact of patent invalidation on subsequent innovation is significant (and large) only in sectors with complex technology” (1). There is however another factor that needs to be taken into account according to Galasso and Shankerman(1): “a high fragmentation of patent ownership among firms”. If these two conditions are met in a particular market, then the patent system becomes a real obstacle to cumulative innovation. However, we cannot say that this is a general trend, given that this seems only to be the case in very specific markets (2).

I can understand that sequential innovations – or even complementary innovations- may be hard to put in place in those specific market mentioned above. Indeed, the time and effort that a second-generation innovator has to put by dealing with all the patent holders, plus the the costs of separate licensing makes the new product difficult to launch given its substantial costs.

But I think we need to see the bigger picture here, and consider another factor that tilt the balance toward second-generation innovations. Once a new invention is commercialized, and is based on previous technologies, it has the advantage of having a new added value compared to products based on “older technology”. Indeed, it becomes complicated for the older products to follow up the new invention in the market. That is exactly what happened to EMI when they developed and patented their x-ray technology. This technology inspired other subsequent inventions that very soon dominated the market, forcing EMI to get out of it (3). So we can’t just affirm that the patent system deters cumulative innovation, because spillovers allow to generate valuable second-generation innovations that heavily outweighs first-generation innovations in the market.

My point is therefore the following: anticommons are indeed a negative effect on cumulative innovation, but I think it is part of the game. Second generation innovation must cope with that in the first place, but then they will enjoy a non negligible advantage compared to first generation innovations.

To conclude, anticommons are a genuine issue that appear in very specific markets, and the patent system has a clear opposite effect concerning the cumulative innovations according to me.

References:
(1) https://www.ipdigit.eu/2014/11/do-patent-rights-facilitate-or-impede-cumulative-innovation/
(2)http://voxeu.org/article/do-patent-rights-impede-follow-innovation
(3)http://voxeu.org/article/cumulative-innovation-and-market-value-evidence-patent-citations

Patent rights are performing a dual role on innovations. While granting an exclusivity right to the innovator, the publication of the invention makes the invention available to the public. These two roles might lead to opposite effects on cumulative innovations, which are defined as innovations based on already existing ones. Problems may occur when the invention requires previously patented innovations, leading to negotiation costs. The patented inventions appear then not to be that available as contended by the so-called “disclosure”. Furthermore, the exclusivity right granted to the patentee allows him not only to exclude competitors to use the invention, but it also excludes second stage innovators to benefit from it.

As J.L. Ferreira underlines it in his article (How patent rights affect cumulative innovation), it becomes actually a significant issue when a large amount of patented resources is needed, leading eventually to abandoned opportunities. If not abandoned, second stage innovators would have to catch these opportunities by circumventing the patented inventions, leading to inefficiency (waste of time and resources). Moreover, this issue formalized in the tragedy of the anti-commons (reflecting an inefficient use of resources due to exclusivity rights) is expected to grow and occur more frequently as the number and specificity of patented inventions increases.

As a result, there is a need of finding an equilibrium between these two effects: How to reward the initial innovator, without excluding others from taking advantage of the patented invention? I do believe a solution could be to raise awareness among first-stage innovators, emphasising that their invention could be a potential source for other inventions and therefore source of value. It would stimulate them to find solutions to benefit fully from their initial invention by making sure it is used for subsequent inventions developed by themselves or developed by second generation innovators.

References:

Belleflamme, P. (2014, November 20). Do patent rights facilitate or impede cumulative innovation? Retrieved from https://www.ipdigit.eu/2014/11/do-patent-rights-facilitate-or-impede-cumulative-innovation/

Ferreira, J.L. (2015, April 17). How patent rights affect cumulative innovation. Retrieved from http://mappingignorance.org/2015/04/17/how-patent-rights-affect-cumulative-innovation/

The main goal of this comment is to briefly investigate how common anticommons are. Furthermore, it is possible to formulate some relevant observations and arguments regarding the influence of patent rights on cumulative innovation.

Anticommons, or the extreme sharing of a resource with the right to prevent others from using it, are a serious and debatable issue when it comes to cumulative innovation. Indeed, the simple fact of making sure that every rights holder is compensated properly can be a tremendous task. Moreover, such a configuration can impede innovation by a) making it costly both in monetary terms and in efforts b) plainly and simply prohibiting the use of the patented innovation.

A good example of this is found in Biomedical Research where two main issues are encountered : a) creating too many fragments of IP, b) permitting too many upstream patents owners to stack licenses on top of the future discoveries. (1) A striking example of the second cause is found in the aircraft production industry during the second decade of the twentieth century: “[…]it was almost impossible to build an airplane in the United States. That was the result of a chaotic legal battle among the dozens of companies—including one owned by Orville Wright—that held patents on the various components that made a plane go. No one could manufacture aircraft without fear of being hauled into court. “ (2) The solution found by the government, motivated by the imperatives of WWI, was to create a “patent pool”. The overall consensus in the literature seems to be that anticommons are extremely common accross all industries

A major issue with anticommons, as stated above, is the difficulty for an “innovator” to assemble all the rights needed to legally take advantage of the innovations protected by the patents. This undoubtedly limits the number of companies able to effectively innovate cumulatively.

A very empirical, and somewhat naïve, example that patents aren’t necessary for innovation to take place is that progress occurred for centuries well before the introduction of any patent system. A somewhat more practical example backing this assumption up is that, in many cases, intellectual property can only be obtained via patent under more expensive terms than would have been available had you just to develop it yourself. (3) Finally, a theoretical point of view, observed by Economic historian Eric Schiff, is that countries that got rid of their patent system actually found that it increased innovation. (4)

As a last example, which still needs to prove itself, I’d like to cite Tesla Motors. Tesla indeed made all of its patents available to other companies, thus effectively allowing them to a) use said patents to produce their own electric cars b) innovate using said patents as premises for new research. This unusual way of proceeding will yield some interesting results in the upcoming years, both in the car industry, but also in the innovation around electricity storage. (5) This approach closely resembles the one used in “Open-Source” software, allowing multiple parties to innovate around the same core elements.

As a conclusion to this comment, it seems difficult to say whether patents increase or impede (cumulative) innovation, but they certainly affect the distribution of innovation across the industry.

(1) http://www.sciencemag.org/content/280/5364/698.full
(2) http://www.newyorker.com/magazine/2008/08/11/the-permission-problem
(3) https://www.reddit.com/r/Economics/comments/6vpgz/the_permission_problem_the_tragedy_of_the/
(4) https://mises.org/blog/patents-and-innovation-2008
(5) https://www.law.berkeley.edu/files/Tesla_Motors_IP_Open_Innovation_and_the_Carbon_Crisis_-_Matthew_Rimmer.pdf

As explained by Professor Belleflamme in this article, the tragedy of anticommons refers to the fact that some IP resources may be under-used because of the exclusivity right it confers to its owner, preventing others to use the innovation. Some cumulative innovations cannot thus be commercialized. Indeed, the licensing costs of the patents you need to put your product on the market become deterring. This, as well explained in the comments below, can be further observed in industries needing a high level of technology and knowledge such as the biomedical and high-tech ones.

This phenomenon, of course, leads to question the positive impacts of patents. Indeed, one the main aim of patents is to allow second-generation innovations to be made thanks to the publication of the first-generation ones, avoiding at the same time the inefficient cost of making twice the research for the first innovation. However, the exclusive right of the patent owner, which allows an innovator to capture the benefits of his discovery (and so is essential to give incentive to the innovator to conduct research), is working against this. This creates a great dilemma that regulators still have to solve: How can we promote further innovation on the basis of other ones without restraining too much the rights and gain of the initial innovators?

For now, the patent system as it is prevents a lot of second-generation products from being put on the market. However, Richard A. Epstein and Bruce N. Kuhlik state that innovators can also invent around an innovation, and that this is not a « wasteful activity » since it will « expand the expand the options available to others whether that invention is patented or put into the public domain » (Epstein & Kuhllik, 2004, pp. 6). Of course, it would be optimal not to have the research done twice but, as it is impossible if the second-innovator wants to be able to commercialize his product, it is better than not granting any patent at all. Indeed, this will result in even less knowledge transmission. Firms, if they don’t get patents, will then try to keep their inventions secret and the access to them will be greatly reduced, or even worth, there will be a lot less research conducted, as the incentives would have been withdrawn.

As a final point, I can add that an other solution to anticommons could be the cross-licensing and pooling agreements (Epstein & Kuhllik, 2004). These would contribute to a wider spreading of patents’ use to several different companies.

References:

Belleflamme, P. (2014, November 20). Do patent rights facilitate or impede cumulative innovation ?. Retrieved from https://www.ipdigit.eu/2014/11/do-patent-rights-facilitate-or-impede-cumulative-innovation/

Epstein, R. A. ; & Kuhlik, B. N. (2004). Navigating the Anticommons for Pharmaceutical Patents: Steady the Course on Hatch-Waxman. Unpublished, University of Chicago, Chicago.

The tragedy of the anticommons appears to be a serious concern regarding the implementation of a patent system within sectors that highly rely on cumulative innovation. For instance, biomedicine is one of the fields where research is heavily dependent on the innovators’ access to the state of the art, to the common pool of scientific knowledge already acquired [1]. Long (2000) states that the current models of intellectual property rights protection do not take into account that in some research processes patents can be granted at multiple stages. As a consequence, these models are not able to alleviate the tension between a patentholder’s interest in maximizing the profit streams from the patented innovation and the public/private interest in allowing downstream research, especially in the case where innovations have both basic and applied uses (e.g. a specific drug that can be marketed by itself, but that can also serve as a starting research point for a new innovation). This is essentially the problem: the static vs. dynamic approach discussion comes into play again. If the decision of granting a patent fails to take into account the potential effect it can have on prospective follow-up innovations/lines of research, then it is not surprising that in industries where innovation is cumulative, patents are perceived as ‘harmful’. Coming back to the biomedicine example, it is overbroad patents to basic innovations that are particularly known to hinder incentives in the downstream research level. Therefore, the question is not only whether a patent in itself can impede cumulative innovation, but also whether there are specific patents characteristics that seem to foster this negative outcome.
Is there any empirical evidence for such a negative effect in the biomedical field? As mentioned above Galasso and Schankerman (2013) indentify a number of difficulties that arise when empirically studying such effects. Williams (2013) investigates whether IP on the human genome influenced subsequent scientific research and product development, and takes into account a specific patent-holder, i.e Celera. The researcher finds evidence that Celera’s IP led to a 20-30% reduction in subsequent scientific research and product development. As part of the methodology, Williams (2013) compares subsequent innovation outcomes for the genes protected with an IP and for non-protected genes that were sequenced in the same year. This procedure overcomes the first problem identified by Galasso and Schankerman (2013). But could give rise to others such as selection bias – namely Celera’s IP-protected genes might have had a lower potential for follow-on research, thus reflecting a lower level of subsequent innovation outcomes. Another potential problem could be that the estimates might reflect a substitution of innovative effort, namely that when facing restricted access to research input, researchers might respond by a shift in their project choices. This can give rise to a new line of inquiry: if the absence of IP is a factor in the decision making process for the line of research of a scientist, the welfare cost of implementing IP should take into account the distortion created because suboptimal projects are chosen [3]. It seems that in this field, more and more questions are to be asked. Even if important steps have been made towards formally proving that IP rights are detrimental for cumulative innovation in some sectors, there are still concerns regarding methodology or robustness.
[1] Long, C. (2000). Patents and cumulative innovation. Washington University Journal of Law & Policy,2.
[2] Galasso & Schankerman.(2013). Do patent rights impede follow-on innovation? CEPR’s Policy Portal
[3] Williams, H. (2013). Intellectual Property Rights and Innovation: Evidence from the Human Genome. Journal of Political Economy, 121(1), p. 1-27

In this comment, we were asked to investigate how common anti commons are and to find other arguments regarding the effect that patent rights have on cumulative innovations.

I would say that, in current times, a lot of companies produce finished goods that have been brought to life after a series of cumulative improvements. If we take the example of the iPhone, we can clearly see it. The iPhone 6s is actually the iphone 1 which has undergone many improvements over time. In a previous article on the IIdigIT, called ” The smartphone patent wars: nothing really surprising…”, Mister Belleflamme identify two types of cumulativeness in the field of cumulative innovations. The two type of cumulativeness are the complementarity and the sequentiality. The complementarity is the fact that you have a new generation product that results of a combination of previous generation innovations. The problem with the complementary innovations is that if many patents have been granted for the previous generation of the innovation, you will not be able to use the previous generation as “inputs” for the further generations. The allocations of too many separate property rights will have higher prices as a result. We call this “the tragedy of the anti-commons”. The other form of cumulativeness is the Sequentiality. It is the fact that a first-generation innovation leads to further generation innovations. The problem is that the first generation innovation confers to the patent holder the same rights over following innovations.

In my opinion, it is clear that the patent protection can impede the cumulative innovation. When firms try to launch new innovations, they must be extremely careful because they can’t use innovations that have been patented without having the licenses to do so! And we know that very little products are completely new innovations. New products include very often innovations that have already been patented and disclosed to the public. It means that companies, when making new product, have to be sure that they can obtain the license they need for each innovation they used in their new product otherwise they will be sue for infringement. All the companies do not have the same budget that they can allow to obtain the licenses they need. This is why, in my view, the cumulative innovation can be deterred by the patent protection.

Obviously, all the companies are not equal when facing this problem. Bigger firms may face less problems than small firms to obtain those licenses but I think that it also depends on the sector in which the firms are operating. A concrete example is the firms that produce high-tech products. If we take the example of a computer, we know that firms have to respect some standards. If they have to obtain a high number of patents when making a new device, it would certainly reduce the incentives they have to innovate. I talked about the field of products but the same conclusions can potentially be applied to service companies.

I will conclude that patents can have different effects on the innovation. We saw during the lectures that the patents can be seen as an incentive to innovate because they confer a right of exclusion to the patent holder, but also because after receiving a patent, you have to disclose your innovation so the others can see how you did and try to innovate too. However, we can say that when it comes to cumulative innovations, patents can have negative effect.

https://www.ipdigit.eu/2013/11/the-smartphone-patent-wars-nothing-really-surprising/
http://www.digitopoly.org/2014/07/17/do-patents-stifle-cumulative-innovation/

“An anticommons property is a scarce resource in which multiple owners have the right to individually exclude others from its use, and no one has an effective privilege of use. Stalemate results in the tragedy of the anticommons, the underuse of a property.” (1) This tragedy has been a problem for the development of our economies for a long time now. From the examples of the past and the present, it is possible to learn a lot about this problem, and how it should be solved.

Back in the beginning of the 20th century, because of all the patents for all the components, it was almost impossible to build an airplane in America. It only became doable after the WW1, because something needed to be done to protect the USA (2). In this case, the intervention of the congress putted an end to that problem, but it surely can’t be the solution to all the cumulative innovations problems. What would be the incentive to patent your innovation if you knew that the authorities could strip you away from your protection as soon as it could actually bring you a lot of money? State intervention is therefore not the best and only solution to that issue; it needs to be associated with other measures.

In the bio industry, the increasing number of patents might have actually slowed down the innovation. Anticommons are therefore very common in this industry, and it can therefore keep people from creating new drugs, etc. In this case, state intervention is no good solution as well, because it would be in the most cases difficult to call it an emergency. In my view, the state could still play a role in this problem, by facilitating the negotiation between the patent holders and the new innovators. It could be a good move from the state to organize some kind of “patent packages” with the help of the patent holders. Such a measure could lower the transaction cost for the new product and make the procedure easier for everyone! I also strongly believe it would be profitable for the patent holders: in some cases, the procedure and the cost of launching a new product is so long and so difficult that a lot of innovators just give up. Sure the innovation remains protected for the patent holders, but on the other hand if it is an unused patent, it could be a good move for him to sell his rights! Such a package would encourage innovators to pay for the right to use patents, and therefore be profitable for everyone.

In conclusion, we can say that patent rights have a lot of negative effects on cumulative innovation, through all the anticommons situations that it creates, but we must remember that this is only the case in some industries. In others, cumulative innovation only requires a few patent rights and thus remains possible. Therefore, there is not a single solution that could apply to all the cases. Like always, the legislator should differentiate its action according to each industry so that innovation can foster while remaining protected through patents.

Sources:
(1) http://www.cooperationcommons.com/node/412
(2) http://www.newyorker.com/magazine/2008/08/11/the-permission-problem

We could take the example of the biopharmaceutical industry. This is an industry that has seen substantive horizontal and vertical integration. Hence, the roles of patents becomes very important in this sector, moreso as the industry relies heavily on cumulative innovation. However, it is necessary to create a fine balance between the economic benefits for a company in the sector and the broad social benefits that accrue from innovations in this sector. In simpler words, there is a need to preserve competition in this sector along with ensuring sufficient benefits for indulging in innovation.

Peter Ringrose, teh R&D chief at Brystol Meyers Squibb once commenetd on the difficulties faced by his firm for not being able to secure licensing rights from upstream patent holders resulting in significant delays for several projects. This is one example which shows that having patents might lead to complications in teh case of cumulative innovation.

Again, the Cohen Boyer method of introduction of genes into host cells was owned by an academic institution. They licensed it at reasonable rates ensuring several agencies involved in DNA recombination technology had access to it. The same might not have been possible had a single vertically integrated firm owned the rights(it might have reserved the rights just for itself)

So, concluding, I would like to state that patents even in the same sector may have varied effects on innovation. It often depends upon the kind of patents, the kind of players involved and the nature of the specfic technology in question.

Patent rights, or more generally all IP, were created and implemented as a way to reward the investment in innovation. This strategy to encourage inventors or companies to invest in R&D may not be perfect, (as discussed in many other articles of this blog) but suggesting that the real effect is the opposite is putting into question the hole IP system. However, cumulative innovation is still one of the weaknesses of the actual patenting model.
This problem applies when the rewards of a first innovation (in a economic perspective, the returns don’t cover the costs) are not enough to incentive the inventor to innovate. In this situation there are maybe 3 solutions:
– The state partially subsidizes research, usually by offering grants or prizes (Should we praise the prizes?);
– Enlarge the patent breadth giving partial rights to the first patent owner over subsequent innovation;
– Wait for the market to develop the right conditions to the innovation to occur. Either improving efficiency (reducing the cost to innovate) or reaching enough capital to incur both first and second innovation in the same company.
It is possible to find examples of this 3 different types of solutions but none of them is perfectly efficient in the society point of view. Since it is the one directly related to the patent legislation we’ll analyze more deeply the second one.

In this case where the rewards of a first innovation don’t pay the costs, the idea of enlarging the rights to subsequent innovations seems logic, and licensing the respective patent would be the natural reaction in the markets. The reality however, is much more complicated. Making an analogy, if everyone may show interest in baking a cake, the problems appear in the time to divide it. Due the specific characteristics of knowledge it is hard to until where the rights of the first patent cover the second one. A narrow patent will not give enough incentives for the first innovator and a too broad one will neutralize the incentives for the second. A midterm is often what is looked for, but the optimal patent design depends on very specific characteristics of each market or situation and cannot be generalized. For that reason, the patent system fails sometimes in leaving enough incentives for the subsequent innovations, leading to innovation “hold-ups” and patent thickets; this situation is often referenced as the “tragedy of the anti commons”. The examples given in the article are very clear showing the retard in innovation caused by this solution (“ the removal of patent protection leads to about a 50% increase in subsequent citations to the focal patent “).

To conclude and state my personal position in this issue I would say that patent system is the best imperfect solution in the most of the situations. To improve its success, it is important to assure the cooperation between later and previous innovators, either by licensing or by mutual agreements. As I also defended in previous comments (Is R&D cooperation a steppingstone to collusion?) the R&D cooperation could be encouraged in order to redirect litigation expenditures to innovation, if the consequences in the competitiveness of the market are not specially large (collusion is likely to occur). Here in this context the same cooperation can work positively in enhancing the development either by allow gin a cooperative group to do both first and second innovation (by sharing costs) or by allowing “defensive patent aggregations”. The state can also have an active roll as stated in the article The smartphone patent wars….

Nowadays, there is no doubt that cumulative innovation is central to economic growth. As far as we can see the environment evolution had changed thanks to multiples cumulative innovations.

There exist different types for these innovations which are based on older(s):
– Sequential innovations refer to the fact that a particular innovation leads to second- generation innovations. The inventor of glass induced several new innovations (such as window, sunglass or watch).
– Complementary innovations refer to a second-generation product which requires the input of a number of different first-generation innovations (such as most of the current new technologies).

Through that, it could be understanding that the entrepreneurs spending times and moneys in R&D to obtain such innovative products/ services want to protect it. That could induce the increasing proliferation of patents and the fragmentation of ownership rights among firms raise transaction costs, constrain the freedom of action to conduct research and development, and expose firm to ex-post holdup through patent litigation.

Overall, few innovations are right now first to lead researches without refer to former’s ones, but it is extremely rare and I will not take into account in my comment.

After that, when I think about the opportunity for the first innovator to protect his R&D, I totally agree but I assume that this protection should be restricted by a kind of patent policy. Because, without this body to control the conflicts among first and second innovator, the opportunity to be the first led to bad faith and abuses (such as patent troll).

But when I look further in the history and the agricultural Indian case, I see that a big company (Monsanto) had patented some seeds. Considered as the first to find them, each one who used them had to pay a license. Fortunately, the Indian government revoked these patents. Today, I see that US patent policy is also trying to avoid this infringement of this common resource. Through this example, I believe that an improvement is possible but too slow.

In order to allow the most efficient trade-off between upstream and downstream, no one should decide if license is needed (and the price) but it has to be regarded more scrutiny by a patent policy officers.

Nowadays, we can say that all innovation comes from a previous innovation. It is really hard to find something that has been totally invented on basis of nothing during the last ten years. According to that, it would be exagerated to say that patents are totally blocking the innovation because new products are created every month by using technology and designs of a lot of previous innovations. The only problem is to find good contracts between all the firms involved in a new innovation, so in other words between the firms using the patents for the new innovation and the firms owning the patents of the previous innovations. An explicit agreement between the first and the second innovator is the most beneficial way to manage the issue of patent. Allowing another to use your patent to create a new product through an agreement is better than blocking the innovation by asking for the real price for the patent. It is more interesting to share the benefits with the newcomer than to no longer have a product using our patent on the market. For example, Microsoft and Intel have worked together for a long time now, and they have shared many innovation. We can still find citation of Intel’s patents in Microsoft’s ones and vice-versa.

We can distinguish the external patents and internal patents. Internal patents are patents that are owned by the firm using them to make the new innovation, and external patents are patents from external firms used by the innovator.

According to a study realized on about 1,200 US patenting firms between 1981 and 1997, it is shown that there is a positive relationship between market value and internal citations of patents. So firm using their own patents to create new innovation are raising their value. On the other hand there is a negative relationship between external citations of patents and market value. So we can say that external patents are restricting the development of the market.

In conclusion, patents are not impeding the incentive to innovate but they makes obstacle to the gain of value of the market in which we use them.

Sources:
http://www.sciencedirect.com/science/article/pii/S0167718705000640
http://www.voxeu.org/article/cumulative-innovation-and-market-value-evidence-patent-citations

“Until a short time ago, the tragedy of the commons was the only tragedy
in town. However, Michael Heller recently introduced the concept of
the “tragedy of the anticommons,” and has systematically explicated its
effect.The tragedy of the anticommons is, in many ways, the mirror image
of the tragedy of the commons. Anticommons property exists where
multiple owners have a right to exclude others from a scarce resource, and
no one has an effective privilege of use.” [1]

If we take a quick look at the litterature, we can immediatly acknowledge that the Tragedy of the Anticommon is an (in)famous effect of the patent system. As explained in the article of Pr. Belleflamme, they are particulary present in sectors like electronics, communications and bio-medicals (genes).

Is the Tragedy of the Anticommons common ?
Though some industries are really hurt by this Tragedy, we don’t think this is a general effect on all the markets, as Galasso suggested : “Overall, our findings show that patent rights block cumulative innovation only in very specific environments” [2]

Other empircal/theoritical arguments over the effects the patent system have on cumulative innovations ?
The reader might want to take a look on my third reference ( [3] ). The author seems to have read Galasso and Schankerman and presents the work of Sampat and Williams over the effect of the patent system on follow-on innovations.
Let me draw your attention on those presice lines :

Prior to this paper, had you asked the 200 economists in the room at the NBER what they thought the outcome would have been, it is fair to say, all of them (including myself) would have predicted that patents would have a negative impact of at least 10 percent. This probably included the authors. As it turned out, the paper showed that those magnitudes in reduction could be rejected statistically. But more to the point, the paper presents pretty convincing evidence that you cannot reject zero as the likely prediction. That is, the effect patents on follow-on research appears to be non-existent

[…]

If we actually look at the economic theory literature on cumulative innovation and patenting, especially the work of Suzanne Scotchmer, we would find that literature predicts that patent rights holders over pioneer innovations will find a way to ensure that follow-on innovators do their innovations and get paid enough to make it worthwhile. The reason is that it is mutually beneficial for that follow-on innovation to occur and in typical Coasian fashion, it works out. To be sure, the follow-on innovators would prefer not to pay the pioneer patent holder but that is secondary to the issue of whether the innovation actually takes place. The Sampat-Williams evidence suggests that for most patent holders, the pioneer innovations do not end up blocking follow-on innovation — at least not because they happen to have a patent.

In my mind, this suggests that as a matter of economic modelling, we would do well to abandon as a default our assumption that innovative effort and investment are inherently non-contractible. Somehow, despite formidable measurement issues, there is some way to work out how hard it is (or was) to come up with the follow-on innovation and to take that into account.

The paper of Sampat and Williams is available here : http://conference.nber.org/confer/2014/SI2014/PRINN/Sampat_Williams.pdf

In short words, the conclusion of Sampat & Williams seems to be that we cannot deny a zero effect of the patent system over cumulative innovations. Therefore, I am personnaly mitigated over the effect of the patent system over cumulative innovations. Both the arguments of Galasso and William seem relevant to me.

Source :

[1] Dan Hunter, “Cyberspace as Place and the Tragedy of the Digital Anticommons” , 91 Cal. L. Rev. 439 (2003).
Available at: http://scholarship.law.berkeley.edu/californialawreview/vol91/iss2/4

[2] Galasso, Schankerman, “Do patents rights impede follow on innovation ?”, 14 may 2013, consulted online at : http://www.voxeu.org/article/do-patent-rights-impede-follow-innovation

[3] http://www.digitopoly.org/2014/07/17/do-patents-stifle-cumulative-innovation/

Well, first of all, I will just come back to the patent market. Encourage innovators to collaborate with appropriate partners can already lead to patent sharing, but of course with royalties. If governments try to encourage patent sharing (by giving subsidies for example, helping firms to pay their royaltier), then the problem for cumulative firms can be partially avoided. So this gives an additionnal incentive to the governments in order to help the patent market.

So, speaking about the tragedy of the anticommon, what we can first say is that it depends on the bargain environment in which you are. And then we can speak again about what was said before about patent markets. If the bargain environment is good in your area, of course it will facilitate creation of cumulative innovations with patents getting invalidated. So this can also explain the 50% increase. Patents are so responsible for a lot of issues but not only.

But, time to get to the point. Speaking about links between cumulative innovations and patents, we can also speak about timing issues. Being able to use a patent from another firm/innovator takes time. When you are developping a cumulative innovation, you will have some needs at some moments, and these needs can be required to go to the next step. Being stucked because of the waiting of a patent can lead you to wait a long and so have some cumulative projects taking more time than expected. This timing dimension is one of the reasons that without any patent restriction, cumulative innovations are more frequent. Royalties are cost, but time also !

The, I will add that this kind of collaboration could potentially arouse the interest of the innovator that has the patent needed. This innovator will then know that others firms are developping something with his own innovation. So it can give him incentives to make similar thinks, or just wanting to know more stuff. The firms that would like to acquire the use of a patent has a risk of information discovering from other innovators/firms.

So in conclusion, time and informations are also taken into account in the tragedy of the anticommon, that can give a new dimension (more incentives) for patent markets.

Sources :
– http://www.nber.org/papers/w20269
–

A problem with cumulative innovation is the tragedy of anticommons.
The tragedy of anticommons refers to the idea that if several individuals own rights of exclusion and exercise those rights, they restrict the access and therefore the use of common resources. This Problem is quite common. Already in 1998, Heller in his book “The Tragedy of the Anticommons: Property in the Transition from Marx to Markets » targeted this tragedy. As we will see below the anticommon problem is frequent in certain fields such as computer, communication as well as electronics and medical instrument. Those sectors have to innovate in order to stay competitive on their market. The tragedy is so an counterproductive effect that impede firm to innovate as fast (and as cheap) as they want.

The main result of the Galasso and Schankerman study is: when no patent protections on an innovation exist, this innovation is 50% more used in subsequent innovations. Based on this fact, we can claim that patents on average do not encourage innovation if we consider cumulative innovations.
We can see the impact of invalidation patent on cumulative innovation explained by Galasso and Schankerman to several details.
Firstly, the publicity effect of the court’s decision (Although, this effect appears after two years and depend on the press coverage of each specific court decision).
Secondly, the impact is heterogeneous. Patent invalidation has a large and statistically significant impact on cumulative innovation in certain fields such as computers and communication, electronics and medical instrument including biotechnology. But in other sectors such as in chemical, pharmaceutical or mechanical technology fields there is only a small and statistically insignificant effect. If we search further explanations of heterogeneity in the technology fields, where the impact of patent invalidation is strong, we can point two elements. The first element is the complexity of technology. We can understand the complexity of technology as the number of patents reliable to the new product. The second element is the fragmentation of patent ownership among different firms. Indeed it is easier for the firm of a new product to negotiate only with few firms than with plenty of them. This is consistent with the failure in licensing as the source of the obstacle to cumulatively innovate.
Thirdly, the relation between the size of a company (and thus the scope of the patent portfolio) and the blocking effect show that the impact is entirely driven by the invalidation of patent in the hands of big firms.
In conclusion it’s important to underline the predominance of the environment to explain the blockage effect. The public policy should therefore target these specific environments.

Bibliography:
http://www.sciencemag.org/content/280/5364/698.full
http://www.voxeu.org/article/do-patent-rights-impede-follow-innovation
http://icampus.uclouvain.be/claroline/backends/download.php?url=L1JlYWRpbmdzL1RleHRib29rcy9wYXJ0N25ldy5wZGY%3D&cidReset=true&cidReq=LSMS2041

Complementary innovation is a type of cumulative innovation, in which the second-generation of the innovation is the combination of several different first generation innovations. The main problem for complementary innovation is the “tragedy of anticommons”, which is “the underuse of a resource when multiple owners all have a right to exclude others from a scarce resource and that no one has an effective privilege of use”. (Heller and Eisenberg, 1998). This short comment aims at investigating arguments regarding the effects that patents have on cumulative innovations. I will consider three empirical studies in different sectors and a comparison between copyrights and patents system.

Galasso and Schankerman (2013) found in a recent study that patents are more likely to block follow-on innovation in markets, where new products rely on a considerable number of patentable elements and in which there is a high fragmentation of patent ownership among diverse firms.

The semi-conductor industry is an example of this type of market. Indeed, semi-conductor firms are engaged in complex cumulative innovations and the sector is characterized by rapid technological changes, costly and depreciating investments and the use of a “thicket” of innovation of others. In an empirical study (Hall and Ziedonis, 2000), the authors studied a shift towards stronger patent rights in US legal environment in the early 1980s by interviewing 100 US semiconductor firms during the period after the shift. They looked at how the shift affected innovative activities of firms in this context and the results suggest that the strengthening of patents’ right increase the risk that the holder of the right can block another from using the technology patented. They also reached the conclusion that the legal shift increases patent portfolio races owned by firms due to the fact that firms want to leverage their power in negotiations and to trade the rights to infringe patents with the help of their patents portfolio. The same results can be applied to other complex industries such as computers or electrical equipments.

The second empirical study collected data on the sequencing of the human genome from public Human Genome Project and from the private firm Celera in order to investigate if intellectual property rights on existing technologies hinder subsequent innovations (Williams, 2010). The author found evidence that Celera’s intellectual property rights led to reductions in subsequent scientific research and product developments on the order of 20 to 30 percent, which suggests that Celera’s IP had more negative effects on second generation innovation than if Celera’s genes had been in the public domain.

In the third study, which empirically tests the anti-commons hypothesis (Murray and Stern, 2005), they looked at the dual characteristic of knowledge: it can be both used in a scientific research article or in commercial applications as patent. They studied patent-papers pairs and lead to the conclusion that the granting of intellectual property rights is associated with a significant but modest decline in the knowledge accumulation measured by forward citations. IP may restrict the diffusion of scientific research and the ability of future researchers to “stand on the shoulders of giants”.

Finally, to look at the effect that patent rights have on cumulative innovation, some authors compared the copyright system with the patent system (Bechtold, Buccafusco and Sprigman). While copyright law offers little protection to follow-on innovators, patent system allocates them some property rights. They reached to the conclusion that the copyright regime leads to decreased incentives for follow-on innovation. However, the patent regime may increase follow-on innovation in the first place, which may create hold-up situations once licensing negotiations start.

To conclude, through these different examples, I saw that these patents can lead to an underuse of the resource and especially in specific industry sectors. As a consequence, authorities have to be aware of this problem and design policies that facilitate licensing to promote cumulative innovation without diluting the innovation incentives that patent rights provide. (Galasso and Schankerman, 2013).

References
BECHTOLD, S., BUCCAFUSCO, C., SPRIGMAN, C., “Play First versus Pay First: Exploring Cumulative Innovation in Patent & Copyright Law”

HALL, B., ZIEDONIS, H., (2000), “The Patent Paradox Revisited: An Empirical Study of Patenting in the US Semiconductor Industry, 1979-95” in Rand Journal of Economics, vol. 32:1, pp. 101-128.

HELLER, A., EISENBERG, S., (1998), “Can patents deter innovation? The anticommons in biomedical research”, in Science, vol. 280, pp. 698 – 701.

GALASSO, A., SCHANKERMAN, M., (2013), “Patents and cumulative innovation: causal evidence from the courts”, CEPR Discussion Paper 9458.

MURRAY, F., STERN, S., (2005), “Do formal intellectual property rights hinder the free flow of scientific knowledge? An empirical test of the anti-commons hypothesis”, NBER working paper series

SHAPIRO, C., (2001), “Navigating the Patent Thicket: Cross Licenses, Patent Pools, and Standard Setting”, in Innovation Policy and the Economy, vol. 1, pp 119-150.

WILLIAMS, H., (2010), “Intellectual property rights and innovation: evidence from the human genome”, NBER Working paper series.

Nowadays, to progress, firms have to innovate. But it is impossible to innovate in isolation, a firm has to use previous (separates) inventions in order to create value. For instance, Edison didn’t invent the light bulb. Indeed, Sawyer and Man developed a light bulb with a filament and then Edison found a bamboo fiver which worked better.
This is what we can call: cumulativeness innovations. There are 2 kinds of them; Sequential innovation (one to all. I.e.; Laster leads to CDs, Laser surgery, Spectroscopy…) and complementary innovations (“all” to one).

As I said, nowadays, to progress, firms have to innovate. But it is impossible to innovate in isolation, a firm has to use previous (separates) inventions in order to create value.
If you look at the communication sector, the phone sector more specifically, you could see a lot of different GSMs with different software. Now, everyone around you in the street has IOS or Android. Indeed thanks to cumulative innovation, more specifically, complementary innovations, Apple and Samsung have a naturally monopoly on cell phones market.
So, nowadays, we are more in a ‘’complementary innovations” situation.

But complementary innovations doesn’t let this possibility to each firm. Complementary innovations represent the situation where a number of different first-generation innovation are needed in order to create a second-generation product, to create value. But, prices are higher if they are set by independent patentees rather than jointly (“Royalty-Stacking”) and it may impede innovation.

This is called; the tragedy of the “anticommons”. Indeed, after a recent proliferation of intellectual property rights in biomedical research, and in order to counter-argument the “Tragedy of the commons”, which helps explain why people overuse shared resources, a different tragedy is suggested; an “anticommons”, in which a scarce resource is prone to underuse because to many owners each have a right to exclude others and no one has a privilege of use. Exactly the opposite of the tragedy of the commons, in which, according to Garrett Hardin, people often overuse resources they own in common because they have no incentive to conserve.
In theory, in a world of costless transaction, people could always avoid commons or anticommons tragedies by trading their rights. But in practice, avoiding tragedy requires overcoming transaction cost.
Indeed, in order to innovate, a firm needs to have several patents, owned by separate innovators and because patents are anticommons, the firm need to negotiate a license with each innovator. The problem is that independent pricing leads to suboptimal usage (Royalty-Stacking). Indeed, according to Cournot, it is like “Multiple marginalization” (for producers: if the price of one complement decrease, the demand for others complements will increase (complement + demand increases for the first one) and the solution for the producers is to coordinate decisions in order to make a big united margin).

So, independent pricing is the source of the problem and, in order to fight against higher transaction costs, price “collusion” or “cartels” are desirables. This is called Patent pools (i.e; MPEG, Bluetooth, RFID, etc…); Patents are sold in a package to decrease transaction costs, patent litigation, the total price… But patent pools are difficult to form, may entail inequitable remuneration, may shield invalid patents, etc… and the problem may not be solved.

In conclusion, nowadays, new products appears faster and faster due to cumulative innovation. By using several first-generation innovations, producers constantly innovate. But to use this several innovations, the firm has to negotiate the license to each innovators which increase a lot the transaction costs (Royalty-staking). Patent pools, by reducing all the costs, are a solution in order to decrease these transaction costs. But they are hard to form and may entail remuneration and so, aren’t sure. This is why I think that the system have to change, in one hand, because patent stimulate innovation and is a source of development. But in other hand because patent rights may impede cumulative innovations due to the insecurity of the existence of patent pools and so the increase of the transaction costs (Royalty-Staking)

Sources :
http://www.voxeu.org/article/do-patent-rights-impede-follow-innovation
http://www.sciencemag.org/content/280/5364/698.full#ref-17

We cannot simply say patent rights facilitate or impede cumulative innovation. As we all know, innovation development depends on new technologies and patent rights are the most important elements for stimulating innovation, namely, patent rights facilitate innovation. Nowadays, new products are created faster and faster because of cumulative innovation. Producers constantly innovate new products by referring prior arts. However, producers have to pay a lot for using prior arts due to the fact that prior arts are also under protecting by patent rights. Superficially, parent rights impede cumulative innovation but actually, parent rights prevent low-capacity producers to abuse prior arts. In other words, parent rights decrease similar low-creativity products getting into the market. Unfortunately, big problems would come out, they are monopoly, duopoly and oligopoly such as Apple and Samsung. About a decade ago, there were a lot of different operation system, but now your phone is basically IOS or Android. With innovation cumulating under the effect of patent rights, new technologies in the same industry would more and more concentrative on one or a few of large enterprises. Moreover, telephones are like a kind of necessity which don’t have substitute goods and hence one of them, Apple or Samsung, will certainly have a natural-monopoly on cell phones market. Patents rights can facilitate or impede cumulative innovation but the influence of patent rights relies on the development of intellectual property law. But other than the disadvantages of patent rights, there are two cases about the positive effect of patent rights. One is about Apple, the other is about Microsoft. Every year, Apple launches its new products having a lot of new patents. The latest model can be supposed to be the cumulative innovation of the previous model. This is a kind of improvement. Microsoft is in a similar situation. After a period of time, Microsoft also launches its own new Windows systems. If they didn’t have patent rights to protect their intellectual property, they would never be that successful.

To sum up, patent rights are the greatest system for stimulating innovation but it is not perfect. There is still some problems having to solve, such as “Privateer” and “Overprotection”. It’s unimaginable to consider the world without patent rights. No one has interest to create. No new technology is announced. No new medicine is invented. How do we improve the law of patent rights which becomes our new topic of era?

My opinion is that patent rights impede cummulative innovation in most of cases and do not facilitate this type of innovation in any way. In general in our period I think also that cummulative innovation is the principal innovation and so that maybe patents rights rather than facilitates innovations by making it more profitable makes it slower.
Concerning cummulative innovation, the most simple fact is that to produce something that requires several patents, we clearly are a downstream firm and patents owners are upstream firms. And in this case (execpt some rare exceptions) your « patents providers » are all monopolists. So I think, like for all other normal goods they have incentive to sell less and at higher prices. In this way, patents clearly impede cumulative innovations.
More obvious, the costs of transaction : if I have to pay more to use patents, I also have to pay more to negociate, administrative costs are also higher. And clearly, if these costs are too high, we do not invest in a possible innovation. And more objectionable : if the innovation lost here were important for an other future innovation (and so on), we can say that patents « destroys » innovation like falling dominoes.

Patent regimes usually enhance innovation, as they are seen as incentives for the innovator to recoup the respective R&D investments. Nevertheless, several academics, researchers and policy makers have been arguing that in fact patent rights can be an impediment to foster innovation, specially within cumulative innovations, being formalized in “tragedy of anticommons”. As stated by Heller and Eisenberg, 1998 the raising proliferation of patents and the fragmentation of ownership rights between firms are seemed to increase transaction costs, to harm R&D and “expose firms to ex-post hold-up through patent litigation”. These concerns have been embraced in public debates in US(National Research Council 2004, Federal Trade Commission 2011) and recent decisions by the Supreme Court (e.g. eBay Inc. v. MercExchange, LLC, 547 US 338, 2006) and in European Union through the implementation of a unitary European Patent and European Patent Court (European Commission 2011).

The imminent question of the impact of intellectual properties rights as patents on cumulative innovation is difficult to infer and, thus there is few evidence of that impact. The problem emerges from the difficulty of that impact measurement and of course to identify a good counterfactual, that is what would happen in absence of patent rights in case of cumulative innovations, given that cooperation should play a decisive role. Joshua Gans illustrates clearly that when questioning:” How do you measure an innovation, measure follow-on innovations or applications or, indeed, the quality of all of this? Then you have to find a situation where there was a patent experiment to allow you to examine causality.”

Works from Fiona Murray and Scott Stern, Heidi Williams and Alberto Galasso and Mark Schankerman have argued that patent protection (or other IP changes) might deter cumulative innovation upwards of 30%.

The empirical research conducted by Galasso and Schankerman (2013-2014) (1) through the patent invalidation decisions by the US Court of Appeal for the Federal Circuit (instrumental variable approach) found that loss of patent protection leads to about a 50% increase in subsequent citations to the focal patent, on average, proving that on average patents impede cumulative innovations. Moreover, the impact of patent invalidation on subsequent innovation is highly heterogeneous, in the sense that it has a large and statistically significant impact on cumulative innovations in technological industries – computers and communications, electronics and biotechnology, but it is small and insignificant in chemical, pharmaceutical or mechanical technology. These outcomes can be explained by two sources: complex technology and high fragmentation of patent ownership within several firms.

Also the study conducted by Williams (2) focused on human genome analyses the difference between the human genome project – without property rights and Celera – with intellectual property rights, concluding that Celera IP lead to substantially significant reductions in subsequent research and implicitly on cumulative innovations.

Another study on semiconductors industry (3) have shown that this sector is also highly reliant on patent regimes, especially after 80´s, when that propensity had sharply risen. It is then reasonable to argue that patents hindrance cumulative innovations also in this sector, due to the fact that these firms engage in patent races aiming at reducing concerns about being held by external patent owners and at bargaining access to external technologies on more favourable terms.

On the other hand, regarding the biotechnology and pharmaceutical companies (4) , they tend to collaborate more, since that pharmaceuticals rely on information based on genetic and proteomic that is provided by biotechnology companies. For example to develop drugs for Alzheimer or single nucleotide poly-morphisms (“SNPs”)pharmaceuticals needs biotechnology ´s firms research. In this sense, patents are not impeding cumulative innovation, because if they exist, they are negotiable or they do not harm innovation.

In conclusion, the impact of patenting on cumulative innovations will depend on the industry sector, and also on the economical or strategically interest of firms to cooperate in R&D.

Sources:

(1). Galasso, Alberto; Schankerman, Mark “ Patents and Cumulative Innovation: causal evidence from the courts”, 2014
(2). Williams, Heidi L. “Intellectual Property Rights and Innovation: evidence from the Human Genome”.
(3). Ziedonis, Rosemarie Ham; Hall, Bronwyn H. “The Effects of Strengthening Patent Rights on Firms Engaged in Cumulative Innovation: Insights from the Semiconductor Industry” 2001

(4). Rai, Arti.K “ Fostering Cumulative Innovation in the Biopharmaceutical Industry: the role of patents and antitrust”, 2001 Berkeley Technology Law Journal

http://www.voxeu.org/article/do-patent-rights-impede-follow-innovation
http://www.digitopoly.org/2014/07/17/do-patents-stifle-cumulative-innovation/
http://writtendescription.blogspot.be/2013/04/galasso-schankerman-do-patents-impede.html
http://economistsview.typepad.com/economistsview/2014/07/do-patents-stifle-cumulative-innovation.html