C&EN: BUSINESS - Icing On The Cake

MARC S. REISCH,C&EN NORTHEAST NEWS BUREAU

PAY DIRT Patented research by this 3M employee can lead to royalty income.

3M PHOTO

Until about five years ago, manysmall and midsized chemical companies didn't realize the value of the patents and know-how they had spent dearly to obtain. They do now: They can have their cake and sell it, too.

G. Alan Osan, chemical practice consultant at McKinsey & Co., says companies that work at it can earn up to 10% of operating income from the sale of licenses to their patents and proprietary technology. But few of the companies McKinsey has studied earn more than 0.5% of income from licensing.

Speaking at a recent meeting of the Commercial Development & Marketing Association (CDMA) in Boston, Osan said that "intellectual property is a significantly untapped resource." Only some of the largest firms such as Dow Chemical and DuPont have made real headway in managing their patent estate for greater profits.

Sam Khoury, president of Inavisis, a consulting company that helps firms evaluate their intellectual property, says a chemical company that earns 5% or more of its profits from licensing has a very successful program. However, that level of success is the exception and not the rule, he says.

In the current economic downturn, many small and midsized companies would like to boost revenues, and they are looking to their patent portfolio as a ready source of income. "They have already spent the money on research. Any licensing royalties they receive are incremental profits," Khoury says. Air Products & Chemicals' corporate director of technology partnerships, John C. Tao, says his goal is to have the company make a return on its intellectual assets that would cover 20% of its R&D costs.

Consultant Willy Manfroy of Bornival LLC says that in the past the chemical industry didn't get as much as it could out of technology. The business is capital intensive, and because of high costs, it's difficult to get a return on innovation. Nevertheless, notable exceptions do occur when a breakthrough process can be profitably licensed to competitors, Manfroy says. They include BP's licensing of its acrylonitrile process and Union Carbide's (now Dow's) licensing of its Unipol linear low-density polyethylene process.

"The chemical industry is just catching up to biotech, pharmaceutical, and life sciences firms," which have long counted on licensing to boost revenues and drive business growth, says David Gates, president of management consulting and investment banking firm Gates & Co. For those who try, licensing out technology can provide a good return on investment.

According to Osan, an active company does one deal a month and can earn $10,000 annually per active patent. But, Khoury notes, the royalty fees that patent licensors obtain can run the gamut from 0.1% of sales generated from a patent for a commodity chemical process to 10% for a high-performance chemical application. For gene separation, other biotech, and pharmaceutical licensing deals, royalty rates usually run between 15 and 20%, Khoury says.

"Competition today can't be defined as I've got it and you don't." Today, the new paradigm is, "I've got it, you got it, but I got it first, and I got it cheaper.

MORE TYPICALLY, companies only do one license deal a year, Osan said. He recommended that firms keep an open mind and "bring an entrepreneurial approach" to their R&D-based organizations. Licensing of technology is a two-way street, he pointed out. Firms not only should consider licensing out their technology but also should consider licensing in technology when it makes business sense. "You want to use the best ideas whether from internal or external sources."

When it comes to filing patents and licensing them to others, IBM takes the cake. Though it isn't a chemical company, it does its share of chemical research. Many licensing executives hold the firm up as a shining example of what can be done when intellectual property is well handled. Some note that because of its enormous reach in the computer industry, IBM is motivated to license its technology to avoid the scrutiny of government antitrust regulators.

Whatever its reasons are, the computer mammoth does make a good case for licensing its technology. In 2002, IBM earned $1.1 billion from the sale of intellectual property, including patent royalties and custom development. That's about 15% of its operating income for the year.

It is also about a quarter of the $4.3 billion IBM spent last year on R&D. The firm received 3,411 U.S. patents in 2001--the most recent year for which data are available. Many of IBM's patents and much of its intellectual property relate to software and information technology, where rapid incremental technology change is the name of the game.

Dow does quite well, too. It earned $276 million from licensing patents and technology to others. That's about a quarter of operating income during a difficult year and a quarter of the $1.1 billion it spent on R&D last year. In 2001, the U.S. Patent & Trademark Office (PTO) granted the firm 216 chemical utility patents.

As for DuPont, it earned about $380 million in 2001--the last year it reported complete information--through the licensing and sale of intellectual property. That's roughly equal to 6% of operating income and 24% of the $1.6 billion it spent in 2001 on R&D. In 2001, it registered 418 U.S. patents.

Jay Rappaport, DuPont director of technology licensing, told attendees at the CDMA meeting that DuPont has 20,000 active patents worldwide, of which it uses only a fraction. "Sales and licensing of intellectual property represent significant opportunities to extract value from intellectual property," he said.

In assessing its patent portfolio, the company examines the business value of those patents and may decide to use patents for its own business or to protect its leadership in a particular market. Other patents may support a spin-off or sale of a business, Rappaport said. DuPont Textiles & Interiors, for example, is no longer central to DuPont's long-term plans, so the firm and its technology will soon be sold or spun off.

Selling or licensing a technology is an option, too. Available patents may be posted on a Web-based bulletin board such as Yet2.com--which DuPont, Procter & Gamble, and the consortium that owned it sold in January to consultants QED Intellectual Property. DuPont also posts patents on its own online Technology Bank. Cross-licensing is the way to go when two or more firms can gain equal benefit from one another's technology, Rappaport said.

LICENSE TO DRILL Air Products licenses out its COPE technology, which uses oxygen to increase the capacity of this oil refinery's sulfur-removing equipment.

AIR PRODUCTS PHOTO

By seeking revenue from its intellectual property, Rappaport said, DuPont can extract value from technologies that aren't a strategic fit with the firm or are underutilized. "Licensing can provide real, often high-margin, revenue based on 'sunk' R&D costs," he said.

John R. Dearborn, vice president of technology licensing and catalysts at Dow, says: "We spend a fair amount investing in technology and pursuing patents. If we develop something novel and unique, patenting is a way to protect it."

ON THE OTHER HAND, if another company develops novel technology that Dow can use, and it would cost less to license it than invent a solution, Dow will license-in, Dearborn says. It's done just that for several technologies.

In some cases, Dow acquires a business so it can obtain the technology. One example is the 1996 acquisition of INCA International for its polyethylene terephthalate and polyester technology. In other cases, know-how and patents have formed the basis of joint ventures, such as the Cargill Dow corn-derived polymer joint venture based on Dow expertise in large-scale polymer production, application development, and commercialization.

Dearborn, who was Union Carbide's Unipol Systems vice president before the merger with Dow, says that for a licensing business to succeed, it must offer licensees low-cost economics, offer the leading technology for a large and growing market, and be guaranteed to work. "For that, they pay us," he says. Thus, Dow "shares in the potential value the technology creates, and you don't invest capital. So you are not taking the commercial risk, only the technology risk."

Jeffrey D. Weedman, vice president of external business development and global licensing at Procter & Gamble, acknowledges that the flow of technology "is in both directions." The company develops not only the detergents and other consumer products for which it is best known but also many of the enabling chemical technologies intended to give it a marketing edge.

"Our CEO wants to see half of all innovations come from external sources--either through technology licensing or acquisition," Weedman says. Conversely, the firm also "spends a lot of time taking internal technology external." For instance, last September P&G licensed to Equistar Chemicals its metallocene catalyst technology for making an elastomeric resin using propylene feedstock. The agreement means P&G might have more rapid access to a new soft-feel polymer for housewares, packaging, and other applications.

And the licensing arrangement also gets suppliers, here Equistar, "to support P&G's goals." Weedman says he doesn't mind that competitors will ultimately source the new polymer from Equistar.

IN A WORLD where antitrust regulators keep a close eye on markets, "competition today can't be defined as I've got it and you don't," Weedman says. Today, the new paradigm is, "I've got it, you got it, but I got it first, and I got it cheaper."

He claims that P&G, which received 316 U.S. chemical utility patents in 2001, offers licensees access to its entire patent portfolio. He does acknowledge an internal P&G rule that only allows the firm to license a product five years after invention and three years after market exposure. "But no one pays attention to that," he says.

"It is a matter of what makes sense. We license to suppliers and competitors. They really are all potential customers," Weedman says. And there are no simple templates for how the firm offers licenses. "Sometimes we license, sometimes we will form a joint venture. We do a lot of financial analysis. It's not just about royalties. That would be shortsighted."

Eastman Chemical has also been managing its intellectual property without simple templates. "We do many types of transactions, not just straight licenses," says Mark E. Proffitt, the firm's technology licensing and alliances managing director. Those transactions include joint technical developments with other firms, acquisitions, divestitures, and in-licensing deals.

"There are no set rules," Proffitt says. Strategies differ depending on the company unit, and strategies for individual units change over time. "At times, we might hold some patents to restrict others from practicing them, or we could switch to an offensive approach and monetize patents for their income stream."

Sharon K. Grosh, director of strategic intellectual asset management at 3M, concedes that "licensing is more common in some industries than others." At 3M, the two units that have the most licensing experience are the drug delivery systems division, from which Grosh herself comes, and the pharmaceuticals division.

Drug delivery systems, for instance, have "commercialization partnerships" with firms such as Berlex Laboratories and Ivax Corp. In order to license out even more of 3M's technology in the future, Grosh says she hopes to form more commercialization partnerships between other 3M divisions and outside partners.

A chemical company that earns 5% or more of its profits from licensing has a very successful program, but that is the exception and not the rule.

Like many other firms, 3M uses a stage-gate process to assess the business opportunities and the probability of commercial success for a new product in research. 3M will be integrating licensing options into this process so that it can generate additional revenues from intellectual assets generated during product commercialization. "Licensing technology into adjacent markets is an excellent way to increase return on an R&D investment without affecting our core business," she says.

"We struggle with issues of what to keep and what to license out," Air Products' Tao says. He'd like to see Air Products get to the point where it would consider licensing core as well as noncore technology to others. "You won't make significant revenues--in the hundreds of millions of dollars--if you're only going to license older and noncore technology.

"In certain markets," he explains, "the opportunities are only so big. By enabling competitors, the technology may endure longer, we could enjoy higher revenues, and gain wider market acceptance."

Air Product holds about 1,000 patents in its portfolio, and PTO issued 55 chemical patents to the firm in 2001. Costs to maintain those patents are high, Tao says. If the company protects its patents in major industrialized countries, it can easily spend $200,000 over the 17-year life of each patent. "That means we spend serious money to maintain those patents," he says, to speak nothing of the time and expense of having lawyers monitor and maintain the patents.

However, just because Air Products has a significant patent portfolio doesn't mean all those patents are marketable. He says only 10 to 20% of the firm's patents are marketable.

David Braunstein, vice president of QED Intellectual Property, says when his firm assesses a client's patent estate, about 20% are high-value patents--the crown jewels of the client, including emerging-technology patents. About 60% of patents are of medium value and are used in the company's businesses. The remaining 20% have little commercial value, and firms may decide to donate or drop them and save hundreds of thousands of dollars.

Braunstein, a polymer chemist who headed technology development at Arco Chemical before it was purchased by Lyondell Chemical, says chemical firms now realize they can make money from marketable patents. Some have used QED's Web listing service, yet2.com, to pitch patents available for licensing.

But the money doesn't start to flow as soon as a company decides it is willing to make its patents available to others. "It takes 16 to 24 months to see results," Braunstein says, from the time a company first reviews its portfolio's potential to the completion of a deal. A firm initially needs to line up business unit head, tax officer, and CEO approvals. And then it must also find a buyer and engage in negotiations, he says. But many more firms are doing just that in earnest today than they did just five years ago.