Open Innovation, page 21
FIGURE 8-1
* * *
Ideas and Intellectual Property
Nonetheless, the amount of arm’s-length transactions in patents and licenses is also substantial and growing. The estimated $66 billion in 1996 in U.S. corporate royalty receipts from unaffiliated entities both foreign and domestic has been growing at an estimated 12 percent each year. Individual corporations profited significantly from these receipts. The IBM Corporation reported receiving more than $1.9 billion in royalty payments in 2001. Lucent also received $400 million that same year. Texas Instruments received more than half its net income in such payments during the late 1980s.7 As the overall market size suggests, and as individual companies have found, there can be big money in licensing one’s IP.
Although big money is at stake, the management of IP seems to have substantial room for improvement. According to a survey conducted in 1998, only about 60 percent of patents held by the top patenting firms around the world were utilized in mainstream businesses.8 Many responding companies had hundreds of non performing patents, which were neither used in their own business nor licensed to any other business. As companies learn of the profits of the exemplars just note dand survey their own patent portfolios, they sense that they can do more with their IP than they are currently doing. As will be discussed in the following paragraphs, however, most patents are not worth much. Consequently, there may not be as many valuable patents among the 40 percent of non performing patents as IP management proponents think.
In addition, companies are paying more attention to selling their own IP to others than they are to buying more IP from outsiders. This is a serious oversight. Companies can realize a great deal of value by accessing an external technology, instead of inadvertently reinventing it internally. Both the buying and the selling perspectives are necessary to improve the management of IP.
Strategies for Managing Patents
Patents traditionally played a protective role in business strategy through their legal ability to exclude rivals from using a company’s own technology. Other strategies such as vertical integration—the dominant mode of organizing innovation assets in the twentieth century during the Closed Innovation era—had also defended a company’s business by allowing safe, efficient transfer of specialized knowledge within a close-knit group.9 In this era, patents were valued primarily as a barrier to entry, not as a source of revenue and profit in their own right.
By the 1990s, CEOs and CFOs began viewing patents and other IP as revenue-generating assets that could directly increase a company’s market value. Licensing out one’s own IP during this era elevated patents and other IP assets to the domain of corporate strategy. Businesses with underutilized patent portfolios began taking their IP off the shelf and using it to generate profits. Companies such as Dow Chemical also sorted through their patent portfolio and donated a sizable portion of it to reduce portfolio maintenance costs (primarily filing fees, language translation, and annual renewal fees to cover administrative costs), which could be quite high, and received a tax benefit for doing so.10
However, these maintenance costs are only the tip of the iceberg in the costs of managing IP The darker side of creating additional valuefrom one’s IP is the cost of enforcement: In the United States, 6 percent of patents incurred some form of legal challenge, leading at times to costly judgments.11In the 1990s, awards were often in the $10 million range, although several passed the $100 million mark.12The costs of litigation can truly add up: In the United States, they were estimated to be as much as 25 percent of aggregate R&D costs of U.S. industries.13
Companies would prefer not to pay royalties for IP if they don’t have to, and the thicket of competing claims of dozens or hundreds of patents can create genuine confusion about exactly who owns what. Moreover, the validity of a patent’s claims is not truly known until after it has been tested in court in an infringement suit. For this reason, companies that might be infringing on another company’s patents understandably do not volunteer their money. Indeed, if the IP owner is not engaged in a business activity that uses the IP, the owner may not even become aware of the infringing activities of other firms.
Litigation is only the last step in the process of monitoring, detection, enforcement, and value realization. Preferable outcomes for IP owners usually include reaching a settlement through cross-licensing, alliances, or retroactive royalty fee payments with or from the infringing partner. According to Jeff George, VP of AT&T’s Intellectual Property Management Organization, “when someone infringes on one of our patents, we take action—but that doesn’t necessarily mean litigation. Usually it means negotiating royalties, cross-licensing or even strategic alliances.”14
Where Patents Come From
Most analyses of managing IP start with the patent-issuance process, that is, the stage at which the company has already received a legal patent. Unfortunately, scholars often pay little attention to the process that most R&D organizations go through prior to obtaining an eventual patent. Yet this is where any useful approach to managing IP must start.
Here is a simplified view of the process that results in a patent. The first step is the report of a discovery or an invention by one or more employees in the organization. In some organizations, these reports are termed invention disclosures. Once a discovery or an invention is reported,the organization in which the invention took place (which is the legal owner of the discovery) must decide whether to file a patent on the idea. Sometimes, the idea may be kept as a trade secret, or it may not be protected at all. As discussed in the later section on Intel, publication even may be the best path for the discovery to follow in some cases.
If a decision is made to file a patent, then the inventor must spend time with a patent attorney, who will file the patent claim with the U.S. Patent and Trademark Office. The USPTO reviews the claim and often asks for additional information, such as other relevant prior art, or how the claims of the patent application differ from the claims of prior patents. If the invention is determined to be novel, useful, nonobvious, and adequately explained, the USPTO may then issue the patent. The USPTO estimates that it takes an average of twenty-five months for a patent application to wind its way through to issuance, and the process costs $15,000 to $50,000 per patent, on average, to complete.15
To understand the process of managing patents once they are issued, we must start with a point often overlooked in discussions of managing IP: Most patents are worth very little financially. For example, in studies of patents from six leading U.S. universities, the top 10 percent of those patents accounted for 92 percent of the royalty payments those universities received. Put the other way, 90 percent of patents from these universities accounted for only 8 percent of royalty payments. These results are consistent with other studies of the distribution of payments for patents in universities and from society at large.16These studies also conclude that most patents are worth very little.
Another related, also overlooked point is that it is very, very difficult to know the value of a patent beforehand. Since filing patents is expensive, companies would doubtless prefer to save the costs of filing the worthless ones—but they have no way of knowing which are worthless.
Chapter 4 provides a context to interpret these facts. Technologies acquire economic value when they are taken to market with an effective business model. When research discoveries are driven by scientific inquiry and are not connected to any business purpose, the commercial value of the resulting discoveries will be serendipitous and unforeseeable. Unsurprisingly, most of these discoveries will be worth very little, although a few may be worth a great deal—once they are connected to the market through some viable business model.
The implication from this is that companies should manage IP to enhance and extend their business models and should seek out new business models for discoveries that don’t fit their present models. Research discoveries from within the company should be evaluated not only on their scientific and technical merit, but also on their ability to strengthen the company’s ability to create and capture value in its business. This in turn suggests that companies should educate their R&D personnel on their business model, so that the researchers can understand the potential connections early on in the research process.
In an informal survey of a number of high-technology companies, I found that companies generally do not educate their researchers about the business side of their innovations.17They do little to share their business model with their researchers, and usually locate their R&D personnel away from the people who plan and execute the business strategy.
A more specific finding in the same vein is the way rewards are given to employees who discover patentable ideas within the company. In a company I used to work for, Quantum Corporation, any employee who came up with an idea that the company decided to submit for a patent received $500. He or she got another $1,000 if the patent was subsequently granted by the patent office. The employee also got a plaque, which replicated the cover page of the patent in bronze, if the patent was granted. That was it. There was no assessment of whether or how the invention helped Quantum advance its own business, and all patents were rewarded in the same way.
Nor was this a unique case. At the time of my survey at Xerox, an employee who came up with a patentable idea received $500—period. And if the Xerox employee came up with ten such ideas, he or she not only received ten $500 payments, but also was invited to a dinner with other Xerox inventors who had ten or more patents. Again, there was no discrimination between patents that directly applied to Xerox’s businesses versus those that had no applicability to Xerox. Other companies have similar symbolic observances for a person’s receiving patents, such as hall-of-fame awards. These companies also make no distinction between patents that have direct connection to the business model and those that do not. Only in the cases of IBM and Lucent did companies take any note of the strategic effect of a patent, and this was only recognized in a few cases long after the patent was received. Table 8-1 shows the rewards that other leading technology companies in my survey provided to their inventors.
For comparison, I have included Stanford University’s policy on rewards for its inventors.18 The difference in incentives is striking: Stanford pays no reward for a patent filing, nor does it pay any award for apatent’s being issued. However, Stanford shares with its inventors a sizable percentage of the royalty stream that its patents generate. Interestingly, Stanford also shares a similarly sizable percentage with the academic department that housed the inventor, and the school retains a final third for its own purposes. (The amounts paid by Stanford are net of Stanford’s costs of obtaining the patent and a charge for its costs of operating its technology licensing office.)
These incentives inside the companies in the table (excluding Stanford) are not very large, to say the least. If the enthusiasts of IP are correct in saying that IP is a critical source of value for companies in the twenty-first century, then one might expect these incentives for inventors to be much larger, to spur them to create more IP If, on the other hand, the value of a patent depends primarily on its being commercialized through a business model, then the weak incentives make more sense: The value comes from the party that has a business model to create and capture value from the patent, not from the invention of the patentable technology itself.
This also implies that companies should find ways to search for, and reward, the creation of effective business models that leverage technologies they seek to license. Absent an effective business model, a technology may be worth little indeed. With an identified business model, the owner of IP has a better idea of where to look for potential buyers and some idea of the value of the idea to those buyers. The importance of the business model in managing IP will be illustrated further with Millennium Pharmaceuticals in the next section.
TABLE 8-1
* * *
Informal Survey of Patent Rewards to Inventors by Selected High-Technology Companies
Company Award for Patent Filing Award to Patent Issuance Other Rewards Other Rewards Comment
* * *
HP $1,000 None NA
IBM $1,500 $500 $25,000 Exceptional patents (in hindsight)
Lucent/Bell Labs $500 None $10,000 Strategically important patents (in hindsight)
Microsoft $500 $500 NA
Quantum $500 $1,000 $5,000
$10,000 Plateau awards at 5th, 10th, 15th, and 20th patents
Seagate $500 $1,000 $5,000 Hall of Fame for 10th patent
Sun $500 $2000 NA
Standford University None None 33% net royalties
* * *
Notice that something else is missing in these reward policies. Nowhere in these companies’ reward policies is there any incentive for employees to identify and access useful external IP. This omission wouldbe perfectly understandable if owning the IP were the key to generating value in today’s economy. Then the external technology would be of little importance to a company’s value and so would not warrant any particular incentive to find it.
If, however, accessing external knowledge is also critical to creating and capturing value, then the omission is a mistake. If external technologies can also support and extend a company’s business model, then companies ought to encourage their R&D staff to survey the landscape to identify potential outside technologies. They could even provide a “bounty” to their staff when a promising external technology is identified and brought into the firm. And they should do this survey before launching next year’s internal R&D projects.
Intellectual Property Strategies in Action: Millennium Pharmaceuticals
A few leading companies discussed in this section exemplify Open Innovation principles in action for the management of IP Each company has a logic behind its approach, which is not a logic of control and exclusion, but instead a logic that connects IP to business models and leverages internal and external IP through those models.19
Millennium is a very young company that has catapulted itself into a surprisingly strong position in the pharmaceutical industry. Founded in 1993, the company achieved a market value of more than $11 billion by the end of 2000 and split its stock twice that year.20 Moreover, Millennium achieved this valuation without selling a single product or pharmaceutical compound; all the company’s activities through 2000 involved delivering information and analysis of potential biological compounds and licensing its technologies for doing this analysis.
Millennium is an instructive example of how IP takes on exciting new possibilities when managed in an Open Innovation mind-set. Many companies act as contract research organizations (CROs) that supply information and analysis of biological compounds to pharmaceutical manufacturers. Prior to Millennium, though, most of these CROs lived from research contract to research contract and essentially charged their customers for the time and expenses of their employees. As small organizations with no control over their IP, these CROs had no way to grow out of what is a low-margin business that lacks economies of scale.
One crucial limit for most CROs is that the knowledge generated from their work belongs contractually to the company paying for the research. This is a typical control mentality over IP that characterizes somuch of the Closed Innovation paradigm. Because of the prevalence of this contractual provision, CROs cannot themselves build on or otherwise use the knowledge that they generate from their work.
Millennium started out doing contract research as well.21 How did it escape from the CRO rut of living from contract to contract, with no control over the knowledge that it generated? It did so by creating a powerful technology platform that allowed it to rapidly discover and validate biological targets and chemical compounds, and by using some highly astute deal making with its pharmaceutical customers. I will discuss the technology platform later, but will analyze the deal making here.
Millennium recognized that its customers would use the results of the contract research within the confines of their business models. Millennium exploited the fact that their customers placed little value on knowledge that did not fit these business models. This was the pattern the company established from its first major deal, with Hoffman-LaRoche (now called Roche) in 1994. Millennium agreed to provide Roche with a number of targets (genes or proteins linked to diseases through various tests that both companies agree on in advance) for obesity and Type II diabetes. Roche had strong interests in both areas and was developing a variety of initiatives to treat these health conditions.
However, Roche was not particularly interested in other possible uses of the targets in diseases outside its chosen focus, such as cardiovascular disease. Because of the capabilities Millennium had established with its technology platform, it convinced Roche that it could identify and screen potential targets more effectively and more quickly than rival CROs. Millennium then assigned Roche the rights to those targets within the domains of obesity and Type II diabetes, but retained the residual rights to those targets for other possible diseases.
This arrangement was a good deal for Roche. The company needed additional targets to feed into its business model, and it had decided to focus on obesity and Type II diabetes. Roche had the scientific expertise to convert the most promising targets into drugs. The company had the clinical and regulatory expertise to manage the Food and Drug Administration (FDA) testing and approval process for these drugs. And it had the sales and marketing assets needed to call doctors’ attention to its drugs when they were approved for use. As we saw earlier with Xerox, Roche assigned little value for a technology (here, a specific target) outside the scope of its business model. And giving away the residual rights to these targets in areas of little interest to it, Roche may have gotten a better deal from Millennium than it would have had it insisted on complete control over all possible uses of the targets. Indeed, Millennium probably accepted less money from Roche than it ideally would have liked. Steven Holtzman, Millennium’s chief business officer, commented after the deal, “We gave a little more to Roche because we were younger.”22
