For the half-year to 30 June 2014, the IPKat's regular team is supplemented by contributions from guest bloggers Alberto Bellan, Darren Meale and Nadia Zegze.

Two of our regular Kats are currently on blogging sabbaticals. They are David Brophy and Catherine Lee.

Wednesday, 21 November 2012

Historical look at Open Innovation


Is Open Innovation a new paradigm?  A break with the past?  No, it’s not really a break, it's just that we don’t understand or remember the past very well.  This is James Bessen and Allesandro Nuvolari's main thesis in their recent work which Bessen presented at EPIP 2012.  They argue that one of the reasons we perceive open innovation to be new is that the majority of textbooks and museums perpetuate the myth of the great inventor.  In his presentation, Bessen pointed out that most of the development of the steam engine, subsequent to James Watt, was freely shared.

In the accompanying paper, the two researchers look at other examples.  The power loom was freely shared as revenues from cloth sales dominated potential profits from licensing and sales of machinery.  Another example is that of Bessemer steel, which was initially produced by an imperfect, patented process.  As the technique improved and a patent pool developed, sharing of information within the pool via publications, employee mobility and sharing of patents and know-how, greatly improved the process. In both cases, manufacturing was more profitable than licensing.
Threads of innovation

Bessen and Nuvolari note that open innovation through sharing of knowledge tends to be limited in duration and occur in the earlier phases of the technology.  Rather than the canonical model (where innovators require the legal monopoly of IP to appropriate the returns from innovation), open innovation implies sharing and diffusing information can be profitable.  Like true economists, they argue that in the case of open innovation, the benefits must outweigh the potential harms.  That is, that the choice not to seek patent protection suggests there are economic benefits associated with sharing.

As Bessen and Nuvolari sum up this argument in their paper,
"... these examples make clear that patents are not universally important to innovation. Indeed, many of these technologies developed outside the purview of the patent system. In some cases, such as with aviation, aggressive patenting put an end to a period of extensive knowledge sharing. Yet in other cases, knowledge sharing and patents coexisted, such as the Bessemer patent pool or the early mechanics who would freely share patented inventions with other mechanics but not with manufacturers (Wallace 1978). In other cases, patents may nevertheless be crucial to encouraging innovation. The relationship between innovation, patents and knowledge sharing is a subtle and complex one. The optimal policy needs to apply to the full range of market conditions, technological maturity, etc., in order to encourage both knowledge sharing and proprietary incentives."
This historical perspective on open innovation suggests that our modern approach to innovation may not have the inventive step we thought it did.  However, how do we, as Bessen and Nuvolari suggest, translate these findings into policy? [On the other hand, Merpel asks, what is "Open Innovation"?  It seems to mean different things to different people. Should it be capitalised?]


6 comments:

Anonymous said...

Frankly, I've always regarded "Open Innovation" as a mere buzzword. Some people have always shared their innovative ideas. Others have protected them through IP rights, and yet others through secrecy. And most reasonable people have used a mixture of the three approaches.

This said, I'm less than impressed by the paper: it regurgitates some perceived knowledge of dubious accuracy. So for example, it repeats the old chestnut about the Wright brothers' patent turning an environment of free exchange of ideas into one of aggressive patenting. However, anyone who has studied the early history of aviation knows that early aviators were an obsessively secretive lot, even before Orville and Wilbur, and that patenting also came quite naturally to them. Even more to the point, if there was any such shift towards aggressive patenting in aviation after 1906, as claimed in the paper, it doesn't seem to have harmed its development much: just compare the advances in that field in the twenty previous years and in the twenty following years...

Apart from that, the paper completely ignores the role of secrecy, and treats every invention that was not patented as one which was "openly shared". This is very definitely not the case, with secrecy playing a very important role, especially before the Paris Convention and the globalisation of patents.

Roufousse T. Fairfly said...

Another example is that of Bessemer steel, which was initially produced by an imperfect, patented process. As the technique improved and a patent pool developed, sharing of information within the pool via publications, employee mobility and sharing of patents and know-how, greatly improved the process. In both cases, manufacturing was more profitable than licensing.

The paper does not develop this aspect anew, but essentially regurgitates another one, which is only concerned with the development of steel production in the United States.

A comparison with continental Europe can put this interpretation in doubt.

Bessemer's application was rejected by the Prussian patent office for lack of novelty (!), and his process was therefore available to the Ruhr steel industry. It proved however rather unsuitable for predominantly high-phosphorus ores until Thomas came along. Of the three patents he filed, one essential title (DE12700) survived opposition in a useful form before the newly founded Imperial patent office by no less than nine competitors, who eventually folded and acquired licenses, albeit not without having driven a hard bargain.

The German steel industry was intensively competitive, and depended on patents. Alfred Krupp owed much of his initial fortunes to his monopoly on bandaged railway wheels, and his rivalry with Jakob Mayer of the future Bochumer Verein was legendary. I therefore doubt that the kind of "collaborative invention" postulated in the paper occurred in Germany.

Collaboration can mean other things, such as the 19th century cartel of the "maîtres de forges" who essentially supplied steel to the French railways at the price and conditions they decided amongst themselves. I wonder how innovative they were.

MaxDrei said...

Haven't read the Paper but doesn't it (as ever) come down to the simple binary question, whether you think you can make more money out of your invention by keeping it a secret or by patenting it?

Take packaging. How you package a pint of milk is not something that can remain a secret. But how about something deep inside the hull of a racing sailing yacht (or flying machine)?

Take food. Patent your new fast food tray cos you can't keep it a secret, but what if you've just invented a better souffle? If we go back to Bessemer's time, was it any different for the cooks, chemists and metallurgists that were in practice then?

Of course, back in Bessemer's time there was no ETSI or other technology standards setting bodies. Methinks they make a decisive input to the binary decision. But, otherwise, there's nothing "new" here, is there?

Anonymous said...

Haven't read the Paper but doesn't it (as ever) come down to the simple binary question, whether you think you can make more money out of your invention by keeping it a secret or by patenting it?

You forget the third option (the whole point of the paper, in fact), which is sharing it.

Some people share their inventions out of sheer altruism, but others reckon that they may in fact earn more money that way. It's a legitimate business model.

Anyway, in my opinion, the best use of innovation is using a clever mixture of all three options. Take Google, for instance. On one hand it keeps Android open source, but on the other hand it protects its search algorithms using a mixture of patents and secrecy.

OctopodaTech said...

From what I've seen of 'open innovation' publically funded research often seems to be its focus. In certain technology areas I think open innovation is just a mechanism for large corporates to benefit from government funded research. They file for patents on the subsequent development work they do and take full advantage of the monopoly gained. However the new phase of open innovation has just begun, and perhaps in time will develop into a fair system

Ron said...

It is usually find specific examples to support diametrically opposite views, and patents are no exception.

An example of the negative effects of absence of patent protection, and the influence of outside influences on the exploitation of inventions, can be found in the early electric lamp industry. While Edison's and Swans' respective patents were held valid in the US and UK respectively, in continental Europe they were either held to be invalid or not to cover competitor's products.

Thus in the US and the UK, the respective patentees were able to close down the competition while their patents were in force and supply lamps made to a consistent quality, although Edison's lack of a licence from Swan in the US for Swan's squirted filament technology meant that his lamps could not be not as efficient as Swan's until he bought the company that held the licence in the US. Swan was unable to benefit fully from his patent due to lengthy and expensive litigation resulting in the company only making a profit during the last 6 years of the master lamp patent's term, and the UK government having imposed conditions on private electric power companies that made their formation unattractive, almost strangling the UK's electric lighting industry at birth.

Conversely, in continental Europe the lack of patent protection led to a price cutting war that resulted in no-one making any money and the customer getting cheap lamps of markedly inferior quality until the manufacturers got together and agreed on a price fixing cartel, establishing a de facto monopoly situation for its members as opposed to the de jure situation in the US and the UK.

While this suggests that monopolies can be a good thing, the opposite case can be made out with reference to the semiconductor industry. In an article celebrating the 25th anniversary of the 555 timer, its inventors referred to the free for all attitude of the silicon valley semiconductor companies that existed in the 1970's when no-one bothered about protection of designs and everyone freely copied ideas from everyone else. This was apparently a significant factor in the rapid advances made in integrated circuit technology at that time. I know from personal experience that much the same attitude existed in the UK at that time. Clients such as the Post Office/BT and the MOD always insisted on a second source of manufacture for any custom designs, and to make this possible, you needed to know details of other manufacturers' processes. Engineers were quite happy to supply copies of documents marked confidential to other engineers in other companies so we could all get on with our jobs without getting the approval of their respective legal departments. I doubt that things are like this today!

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':