Comments for Do standard setting organizations’ rules stand in the way of innovation?

Standardization can be harmful for Consumer’s Surplus as a single firm gets more market power and in result of this less diverse products for consumers. I’m going to proove this trough real life examples.

First, as a kid I used to play to video games. I had an XboX and all my friend a PlayStation. Unfortunately we couldn’t share our video games and this was embarassing for me. I had to buy a PS although I prefered an XboX.

Another example I can give is with iPhone users. Almost all the smartphone on the available market requires the same loaders. Their price are very affordable. However, iPhone users must buy a different loader, incompatible with other smarthphones, for their phones. Therefore, Apple can charge prices that can be judged excessive by some of it’s clients.

Also this process can lead to abuse of dominant position. This was the case with Intel, who sells the standard processor for computers and the European commission find them for giving « rebates to computer manufacturers on condition that they bought all, or almost all of their x86 CPUs from Intel » and for « direct payments to OEMs to halt or delay the launch of specific products containing a competitor’s x86 CPUs and to limit the sales channels available to these products »(1).

Sources :

(1) http://ec.europa.eu/competition/publications/cpn/2009_3_5.pdf , page 1, first paragraph.

Winston Churchill once said that democracy is the worst form of Government, except for all others that have been tried. In this comment, I will argue that much of the same might be said about FRAND pledges. Despite inherent weaknesses, such pledges have become the norm to address the market power issues raised by standardization. This is probably due to the absence of strong alternatives.

Though standards can yield vast benefits [1], creating them is not without difficulties. In most cases, high-tech innovations rest on a myriad of patents.[2] As a result, it takes a vast coordination effort to create new technologies. This is where standard setting organizations (SSOs) come into play. Engineers from different firms gather within these SSOs to design the technologies of tomorrow. In doing so, they face a crucial decision: choosing which patented technologies will make up the standard. How these decisions are made can have large market power ramifications.

Imagine a simplified example. A hypothetical industry, comprising eight firms, gets together to create the Video Player of the Future (“VPF”). Three of these companies are competing manufacturers. They don’t own any technology (in the SSO world, they are referred to as “implementers”). The five other firms merely license technology (“licensors”). Of these five licensors, one has a patent on the only viable video format. Another firm has a patent on the only usable lens. Finally, the last three firms offer competing/substitutable disc technologies. To create a fully functional VPF player, each implementer must obtain a license for the lens, the video format, and the disc technology. When they meet within the SSO, the eight firms will thus have to agree which of these technologies they include in the standard.

The lens technology and video format raise the least problems. The owners of these two technologies gain no extra market power from the standardization process. Whatever the SSO does, these licensors will hold patents that are essential to the VPF player. The only issues are Cournot complements (also referred to as Royalty stacking) and potential holdup.[3] First, firms that hold complements each set their monopoly price. The cumulative price is higher than if the complements were held by a single monopolist. Both firms and consumers are left worse-off. Second, standard setting may lead to opportunistic behavior where licensors threaten to litigate implementers in order to extract higher rents.

Choosing the disc technology raises more issues. When the engineers meet, they will have to make a big decision: either leave the disc technology out of the standard or choose a single disc technology.[4] Ideally, the SSO would leave the disc technologies out of the standard. This would maintain the benefits of standardization while ensuring there is competition between the licensors of disc technologies.[5] Unfortunately, this might not always be possible (for example, because the other technologies have to be designed around the disc). The engineers might thus have to decide on a single technology. Doing so removes competition between the disc technologies.

Summarizing, the standardization process leads to two issues: the creation of market power through the standardization agreement, and questions regarding the pricing of standard essential patents (these include royalty stacking and potential hold-up). Though they raise distinct legal questions, both of these problems are usually addressed – at least to some extent – by requiring SSOs to take FRAND pledges from the holders of essential patents. The idea behind FRAND is simple: companies agree that they will charge “Fair Reasonable And Non-Discriminatory” rates in exchange for participation in the standard. This should limit the prices that firms can charge once they are accepted in a standard.

The problem is that enforcing FRAND pledges has proved problematic. One need only glance at other network industries, such as telecommunications and electricity, to see why this is the case. When these industries were first liberalized, most firms were subjected to rate of return regulations, and then price cap regulations.[6] In both cases, regulators ultimately lacked the necessary information to efficiently perform their duties.[7] FRAND pledges raise similar problems. They force antitrust authorities and judges to act as price regulators, even though they lack the relevant information. This problem is even more acute in the SSO context than in traditional network industries. Patents are incredibly idiosyncratic compared to commodities like telecommunications and electricity. Moreover, the very definition of an essential patent is that it no longer has any close substitutes against which its price can be compared. Incentive compatible schemes, where firms’ returns depend on their performance compared to a benchmark, are thus out the question.[8] This has led many scholars to look for alternatives to FRAND pledges.

One such alternative was proposed by Lerner and Tirole. They suggest that firms offer “structured price commitments” before they are selected into a standard.[9] Firms holding substitutable technologies would thus compete for the market. Unfortunately, this solution is far from perfect. For a start, it may be impossible for firms to put a price on their technology years before it actually reaches the market or before the standard materializes. Second, though this solution limits hold-up problems, it does little to address royalty stacking issues (holders of patents without substitutes would have no incentive to offer a low price). Third, this solution might be hard to implement because competition between SSOs could drive firms towards those organizations with laxer rules.[10]

More speculatively, recent research regarding public decision-making might also be relevant. Lalley, Posner & Weyl suggest that Quadratic Voting induces parties to reveal their true valuations in collective decision-making procedures.[11] The basic idea is an auction where parties can buy as many votes as they like. The cost of each vote is the square of the number of votes bought by the party. The winning proposal is the one with the most votes. The money that is collected is divided equally between the voters. Though the authors do not refer to this possibility, SSOs might use quadratic voting to decide whether licensors can price freely or whether the technology should be given away for free. Inducing firms to reveal their true valuations could help alleviate royalty stacking and holdup problems. There are, of course, considerable difficulties. Most notably, it would probably be necessary to modify the money allocation mechanism to ensure that licensors always earn a positive return after the vote. This would affect firms’ incentives during the auction.

Though the Quatratic Voting example is probably closer to science fiction than reality, and Lerner and Tirole’s price commitments would be hard to apply, they illustrate an important point. There is ample scope for mechanisms leading to more efficient outcomes than FRAND pledges. The big challenge is to find such mechanisms.

[1] There are huge benefits to be had if firms can agree upon a single interoperable technology – think of Wifi, 3/4/5G, or Blu-Ray. This is notably due to network effects, which can make fragmentation more harmful than monopoly.

[2] For example, the Blu-Ray player calls upon roughly 8864 standard essential patents. This is the number of patent contained in the “Blue One” patent pool, which contains all the patents that are necessary to produce a Blu–Ray player. See http://www.one-blue.com/license-programs/bd-player-recorder/. Because including non-essential patents in pools can raise antitrust issues, we must assume that a large chunk of these 8864 patents are “standard essential”.

[3] See Mark A Lemley & Carl Shapiro, Patent holdup and royalty stacking, 85 TEX. L. REV. (2006).

[4] Including substitutes (in this case, multiple disc technologies) in the list of patents required to comply with a standard is akin to a cartel. In other words, including substitutes in the list of patents required to comply with a standard turns them into complements.

[5] If this is possible, including disc technology in the standard would be anticompetitive.

[6] JEAN-JACQUES LAFFONT & JEAN TIROLE, COMPETITION IN TELECOMMUNICATIONS 84 (MIT press. 2001).

[7] Id., at 44; 85; and 86.

[8] Id., at. 52.

[9] Josh Lerner & Jean Tirole, Standard-essential patents, NBER WORKING PAPERS (2013).

[10] Id., at 36.

[11] See Eric A Posner & E Glen Weyl, Voting squared: quadratic voting in democratic politics, 68 VAND. L. REV., 475 (2015). See also, Steven P Lalley & E Glen Weyl, Quadratic voting, AVAILABLE AT SSRN 2003531 (2015).

“Do standard setting organizations’ rules stand in the way of innovation?” – I don’t think so

As Sophie Poukens mentioned in her article, there have been these past few years some sort of innovation network growing (i.e. interoperability). Indeed, every time a new product is released, whether cutting edge or existing technology, that product is somehow already connected with existing and/or future items on the market. This is made possible thanks to the standard essential patent and, we must face it, that’s smart. Our civilisation wants to have everything connected for simplicity and range of use. However, blurred rules decided by the SSO make things not as simple as we think it is. In other words, orchestrate essential patents is not as easy as connecting a Bose Sound System to a Sony PlayStation via Bluetooth any other mean. This has come to the question stated earlier.

The first thing I would like to say which is that without rules there wouldn’t be any SSO so this question could be rephrased as : Do SSOs stand in the way of innovation ? The answer to this question is simple giving the elements in the text. Without standards, interoperability would we less effective and wouldn’t lead to such progress.

Having said that, the real problem is the blurred rules around these essential patents and the unbalanced market power. If laws protect consumers against abusive market power there should be laws that protects licensees against abusive cow-boy patent holders and SSOs. Beyond the two conditions for a patent to be essential, in order to be part of the Standardized patent pool, the patent owner should accept unbendable rules, legal rules. His reward being the fact that his patent will be used world wild for innovative technology. If he doesn’t agree, his patent shall remain outside the pool and thus unused by the next gen companies. There should be incentive to encourage standardization but the fact that a patent could be part of the pool is already a big reward regarding the future royalties coming from thousands of companies.

All in all, the pricing rules will be at an equilibrium just like free markets does. At the beginning, the patent holder won’t have any market power (considering it is not essential yet) so negotiations can be handled and price will be taken into consideration by demand. So the only things to take into consideration are R&D investment, return on these and the incentive to standardize it. But the key is to avoid a unbalanced market power. There will be no hold up if the patent holder commits to the initial agreement. It is important to keep in mind that it is the licensees who promote innovation. As a result, they must be protected as a priority such as consumers (who are the base of economic growth) must be protected against monopolists.

Sources

– https://www.youtube.com/watch?v=8a07jTJIZbc
– “LLSM2041 Economics of innovation lecture 10”
– https://www.ipdigit.eu/2015/07/do-standard-setting-organizations-rules-stand-in-the-way-of-innovation/
– http://apps.americanbar.org/antitrust/at-committees/at-ip/pdf/abstracts/20_lerner_efficient_patent_pools.pdf

I’d like to focus attention on the role of the F/RAND commitments in trying to solve the ‘hold up’ and the associated litigation costs as well as that of ‘royalty stacking’ problems.
As emerges in the literature, solutions from ‘within’ the industry are preferable, the SSO in order to codify more credible F/RAND commitments have to reform their IP rights policies. (In line with (1)).
Indeed, I think that, while the subsequent rulings of the courts following litigations can practically define ambiguous F/RAND criteria, too, this process needs to be helped via some internal SSO reforms. For example, to solve the problem of high litigation costs from the challenging party during an ‘hold up’, it is needed a mechanism that blocks ‘cost-raising’ actions by the patent owner, such as exclusion orders, before some precise steps have been taken. Those can be waiting for the evaluations of third parties (as I will talk later). Of course, this kind of mechanisms need to be codified and agreed before the standard is set.
The need of mediating third parties to complement SSO arises because because as Adam Smith said “People of the same trade seldom meet together… but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices”. I think for example that representative of consumers organizations or specialized competition policy agencies (not having in the patent owners their payers) could have a role in SSO. This in order to stay in between the voluntary market-driven SSO and state control as in China.
Another way to reduce the market power leading to ‘hold ups’ is deciding fixed royalties before the lincensees invest (2). All these measures must be directed to avoiding ex-post disagreements and litigation costs.
Finally, the problem of ‘royalty stacking’ is well dealt with by J.Contreras, which proposes pseudo patent pool agreements, in which the patent holders set a collective royalty. Here the previously fragmented owners result incentivized in offering a simple and understandable patent license.

References:
(1) ‘Kuhn, Morton, Shelanski ‘Standard Setting Organization Can help Solve the Standard Essential Patents Licensing Problems’
(2) ‘J.Contreras A pseudo-Pool Approach to Standards-Based Patent Licensing’

Debates upon standard setting organizations’ rules go to two balance weighting processes.
The first one is the balance between the proper reward to the patent owner and the license fee that other potential innovators need to pay to get in this standard, both of which affect innovation. The higher the reward to patent owners, the more incentive to innovation, which however leads to higher cost for the outsiders to use this patent standard, slowing down or decreasing potential innovations. The lower the license fee, the broader this patent standard will spread out across the industry, but which will discourage the initial innovators. The Key to achieve the ideal of fair patent standard lies not only the theoretical idea, but more on the application and regulation of using this patent standard (This turns to the typical question of the optimal patent standard length and breadth). One possible approach to solve this conflict goes to the regulator, that is, to supervise the patent owners’ behavior, to distinguish whether patent owners opportunistically take advantage of its market power to charge unreasonably high license fee to the outsiders or not. By doing so, patent standard insiders will accordingly adjust the price or license fee to a reasonable level to avoid being accused of anti-competitive.
The second one is the balance between the efficiency of innovation and ultimate customers’ well-being. In economic cases, normally, a monopoly or oligopoly is more efficient in exploiting limited sources, especially in newly emerging area. Establishing a standard is good for a group of elite innovators to focus on this specific technology and industry to achieve the highest, thus optimize social resource and avoid wasted energy due to duplication but incompatible work. But after innovating phase, when it goes to the market, competition law should be enforced to protect customers’ benefit. Since concentrated power in a standard by the patent owners help them gain obvious market power, it is easy for them to set price high, which forms a sellers’ market. Then, it is necessary to do the price-cost assessment to judge whether the price goes too far as abusive to the customers. If prices are significant above cost, or the margin is too high, it can be identified as abuse, which should be regulated. Therefore, the question is not whether to set a standard to facilitate innovation or to not set standard on behalf of the customers, but rather the questions of how to set a proper standard to motivate innovations in the innovating phase and to prevent the patent owners from harming the customers after they are actually put into the market.

One of the earlier accounts of a standard setting organization (SSO) dates back to the beginning of the 20th century. The British “Engineering Standards Committee” initially simplified the construction sectors varying measures to increase efficiency in the market. The institution, which later became known as the British Standards Institution in 1931, went on to standardize many other areas in need of harmonization. (1)

Today international standard organisations (ISOs) exist for multiple fields of research, industry and technology on an international level. The International Communication Union for example coordinates the standard setting in the field of telecommunications. Many other such international entities exist to further the implementation of standards and enable national industry standards to be adapted for international trade.

Lately a new branch of standard setting bodies has evolved. Industry consortia, also known as standard setting organisations today, coordinate these endeavours between firms instead of as before between governmental or supranational organisations. Examples of such industry consortia are: World Wide Web Consortium (W3C), which has set standard for CSS, HTML and XML, but also the Bluetooth or RFID (Radio Frequency Identification) SSOs. The latter having provided the industry with patents from multiple firms to ease access to their technology.

Let us examine the benefits of such patent pools known as SSOs. A government release by the United States Patent and Trademark Office details, “patent pools can eliminate the problems associated with blocking patents or stacking licenses […] while at the same time encouraging the cooperative efforts needed to realize the true economic and social benefits of […] inventions” (2). The study issued by the USPTO was concerned with the formation of patent pools in the biotech industry. The conclusions derived from their analysis can, however, be reapplied to different industry sectors, such as information technology. The lowering of legal hurdles, as well as the advantage of lower transaction costs of a firm with a patent pool or SSO, rather than with multiple firms, fosters innovation. Furthermore, possible litigation threats through unknowingly using patents are decreased through the formation of SSOs. Small firms can readily gain access to technologies and thus contribute through their individual innovations to the progress in a specific technological segment. (2)

As risks associated with capital intense R&D spending increase, a standard setting organisation can function as an insurance towards the latter. Patent pools generate a consistent flow of income, if utilized by the industry, and thus incentivize firms to provide their patents. (2) Such an example is provided by the MPEG patent pool/SSO, in which each member gets an equal pay-out and has access to the pools’ patents. This increased form of information exchange within the group of an SSO fosters inter-company development and can boost efficiency as overlapping areas of research are reduced through communication. (2) This practice of information/patent sharing and licensing is more important in capital intensive industries such as the biotech and IT industry.

From a simple competitive aspect, SSOs can lead to diminishing competition through setting standards and thereby eliminating alternatives, or making the implementation of alternatives increasingly harder. However, the lack of standards can lead to wide-spread confusion in markets (which is why the Engineering Standards Committee was founded over a hundred years ago) and create difficulties for consumers who demand compatibility between companies’ products, i.e. the Bluetooth SSO. Ergo, the implementation of SSOs fosters innovation in their specific industry, can lead to increased communication between companies and lowers legal hurdles as well as costs involved with licensing various patents.

(1) Robert C. McWilliam. BSI: The First Hundred Years. 1901-2001. A Century of Achievement. 2001. Thanet Press. London

(2) United States of America. United States Patent and Trademark Office.PATENT POOLS: A SOLUTION TO THE PROBLEM OF ACCESS IN BIOTECHNOLOGY PATENTS? By Jeanne Clark, Joe Piccolo, Brian Stanton, and Karin Tyson. N.p.: United States Patent and Trademark Office, 2000. Print.