CHAPTER 1

THE NEW U.S. PATENT LAW: WHAT IMPACT ON CHEMICAL INVENTIONS?

Adda C. Gogoris, S. Peter Ludwig, and Jason M. Schwent

Darby & Darby, 805 Third Avenue, New York NY 10022.

1 ABSTRACT

On November 29, 1999, President Clinton signed the most important revision of the U.S. patent statute since 1994, and arguably since the original enactment of the statute in 1952. Although the new law is named the American Inventors' Protection Act, it is no protectionist's tool. Its major provisions affect US. patents in more or less the same manner, regardless of the nationality of the patent holder.

In this paper, we briefly outline the major changes in the law and attempt to assess their impact on patents in the chemical field. Of course, an in-depth evaluation of its impact is not possible this early in the life of the new statute since the U.S. Patent & Trademark Office ("USPTO") has not yet passed regulations to implement the law, and the courts have not yet interpreted it in their decisions. Nevertheless, even a preliminary analysis reveals advantages and disadvantages of the new statute, and suggests strategies that can be used to reap some of its benefits and to avoid some of its pitfalls.

2 INTRODUCTION

The following six changes in the U.S. patent law are most important for chemical inventions:

1. Publication of Applications: The vast majority of U.S. patent applications will be published - they will no longer be kept secret until grant of the patent.

2. Liability for Infringement Before Patent Grant: An infringer's liability can begin upon publication of a patent application, i.e. long before a patent is issued.

3. New Defense: A Prior User of a Business Method Avoids Infringement of a Later Patent: Under prior law, the first inventor of a method or product patented by another may nevertheless be liable for infringement. The new law allows the first inventor of a business method only to avoid liability for infringement under certain circumstances.

3. Extension of the Patent Term to Compensate for USPTO Delay: The USPTO will postpone patent expiration to make up for time lost as a result of the USPTO's delay in processing the patent application.

5. Co-Workers' Patents/Patent Applications Are Not Prior Art: Unless patented or published more than one year prior to the filing of a patent application, patent/applications of an inventor's co-workers cannot be used to deny him/her a patent.

6. Contested Patent Reexamination: The challenger of a patent now has the option of participating actively in reexamination of the patent by the USPTO.

The foregoing changes advance two main policies: harmonization of U.S. patent laws with the laws of other countries (changes #1, 2, 3 and 6 above) and greater accountability by both the private and the public sector (#4 and 6). Implementation of these policies ultimately increases certainty in business and decreases the cost of obtaining patents.

The degree to which these policies have been served by different aspects of the new law varies greatly, however, often as a result of political compromise.

But let us examine each of these important changes separately.

3 PUBLICATION OF APPLICATIONS

Until now, U.S. patent applications were maintained as secret until a patent was granted.2 This treatment contrasted with that of most other countries which publish all patent applications 18 months after their filing. In the ever shrinking global community, the vast majority of those who apply for a patent in the United States also seek patent protection internationally. Therefore, secrecy of only the U.S. patent applications arguably afforded these applicants no advantage since the same subject matter became publicly available through publication of the corresponding non-U.S. applications.

A number of U.S. patent applications, however, lack non-U.S. counterparts. Most of these applications are owned by individuals, small companies and universities, although, occasionally, large companies decide to pursue patent protection only in the United States. Such US.-only patent applicants opposed publication of U.S. applications because they perceived that secrecy shielded them for a limited period of time from competitors with greater resources. They argued that, without secrecy, their competitors could gain early access to the basic ideas in their patent application, adapt them quickly beyond the scope of the patent, and dominate the marketplace.

However, the very secrecy thought to be beneficial by these patent holders, also makes it difficult for businesses to evaluate the risk of patent infringement. Businesses (large and small) commonly ask for legal advice on whether their activities will infringe patents owned by their competitors. In response, lawyers identify and study relevant patents, render opinions as to whether an infringement problem exists, and sometimes suggest modification of commercial practices to avoid infringement.

Prior to the present reform, it was often difficult to determine whether any pending U.S. patent applications posed an infringement risk. The only available means for U.S. attorneys to assess the potential risk of infringement presented by pending U.S. applications was to look to published corresponding non-U.S. applications. While lawyers analyzing these non-US. publications were routinely able to make some judgment as to whether a commercial activity might someday infringe a U.S. patent, the analysis had serious limitations.

Analysis of counterpart patents in other countries (when available) often did not provide sufficient information upon which to assess whether a pending U.S. patent application poses an infringement risk. Occasionally, U.S. patent applications have protracted pre-issuance proceedings in the USPTO, which may cause them to remain pending for many years. Since these proceedings are also shrouded in secrecy, it was impossible for a competitor to determine whether a particular U.S. patent application was still pending and whether it was likely to mature into a patent that might be infringed. In addition, patent applications pending outside of the United States often do not correspond to the patent that ultimately issues in the United States as the scope of a U.S. patent need not be identical to that of its non-U.S. counterparts. A company might first obtain a narrow patent, e.g., in Japan, and several years later a broad patent in the US. In addition, patent applicants compounded the uncertainty by employing different patent strategies in different countries.

All of this made it more difficult to evaluate the risk of patent infringement. In extreme cases this uncertainty manifested itself in so-called "submarine patents." Applications for such patents lie in secrecy prior to their grant, eluding the industry's attention. When they "surface'' upon issuance, years after filing and often after competitors have made a substantial investment, they take the industry by surprise. Competitors are forced to pay large sums for licenses of submarine patents in order to protect their investment, or to cease the infringing activity altogether.

U.S. patent policy has been influenced by a recent worldwide trend to reduce uncertainty in business thereby facilitating global business strategies and reducing costs. Beginning with the patent law revision of 1994 (implementing the GATT Uruguay Round Agreement), the U.S. sought to reduce uncertainty with regard to patent infringement. Prior to GATT, US. patents remained in effect for 17 years from the date of issuance, no matter how long the patent took to obtain. After GATT, patents were subject to a maximum 20-year term from the date of filling.

Prior to the GATT reforms, it was impossible to determine when a patent issuing on a pending patent application would expire until the patent issued because any delays or protracted proceedings prior to the grant of the patent would delay the start of the 17-year patent term. After GATT, businesses gained the ability to assess when a patent would expire and to plan their actions accordingly. Patent applicants who sought to extend their patent term by delaying patent issuance were also dealt a major setback. After GATT, applicants who themselves delayed proceedings in the USPTO did so at the expense of their own patent term.

It was hoped that uncertainty would be eliminated by requiring mandatory publication of all U.S. patent applications. However, the position advocated by U.S.-only patent applicants prevailed, and the new law stopped short of this. As a result, some uncertainty remains.

In the simplest terms, the new law requires all U.S. patent applications with corresponding applications in foreign countries, to be published 18 months after their earliest filing date. If a U.S. patent application has no foreign counterpart, then the U.S. application can be exempted from the publication requirement by the filing of an appropriate certification. In addition, if the U.S. application contains subject matter that will not be present in the non-US. patent, that subject matter may be withheld from publication. In other words, if the patent application discusses a body of technology and only a portion of that technology is to be included in both the U.S. and foreign applications, the solely U.S. technology does not have to be published.

At first, this does not seem like an enormous change. All of the applications subject to publication would have been published elsewhere anyway at 18 months from filing. The only real difference is that now, instead of looking to the foreign counterparts of U.S. applications, businesses can look directly to U.S. patent applications, which must be published in the English language. This makes it easier for U.S. patent examiners to search for prior patents and patent applications, and (theoretically at least) will improve the quality of newly issued U.S. patents. This is because the closest prior patent references will be available in the English language and will be more likely to be uncovered and considered by patent examiners.

Additionally, the U.S. publications will have some effects not recognized in their foreign counterparts. In general, U.S. patents are granted for "new" and "useful" invention. A patent is barred if it is shown that someone else has previously created the same or a trivially different invention and placed it in the hands of the public (e.g. by publication or commercialization). Patent examiners search the patent and scientific/technical literature to determine if the invention claimed was discovered by anyone else before the patent applicant. These prior patents and literature are called "prior art." However, not all prior art has the same effect. For instance, non-patent publications (technical literature and non-U.S. published patent applications) are effective as prior art only upon their publication date. Issued U.S. patents on the other hand are effective as prior art upon their filing date which can be several months earlier than the date of publication of a corresponding non-U.S. application, and years before the U.S. patent issues." The new legislation provides that published U.S. applications (presumably even those which do not result in a patent) will constitute prior art references (that can be used to deny a patent to another applicant) as of their (earlier) U.S. filing date," not as of their publication date (which will lag filing by 18 months).

There is yet another, more subtle, reason that the new publication requirement is significant: it breaks the secrecy barrier in the U.S. and opens the door for full unconditional publication of all patent applications and of proceedings before the USPTO, which lead to their issuance ("prosecution history"). This reform opens the way for eventual fully open proceedings before the USPTO in which all patent applications (regardless of whether they are also filed elsewhere) would be open to public inspection. Such a result would further reduce uncertainty regarding potential infringement and would bring U.S. patent law closer to the patent laws of most other countries.

Even with unconditional U.S. publication, and USPTO proceedings fully open to public inspection, one fundamental uncertainty regarding patents would remain. The U.S. still follows a "first-to-invent" approach. Under this thinking, the first person to invent a product is granted the patent, even if he/she is not the first person to file a patent application. The majority of other countries follow a "first-to-file" approach, which rewards the first person who applies for a patent. This difference in approach can have important consequences. In first-to-file countries, it is easy to decide who is entitled to a patent (as between two persons applying for a patent on the same invention) just by comparing filing dates, The first person to file obtains the patent, and the second does not. The patent, once obtained, cannot be challenged on the basis of an unpublished or an unpatented prior invention by another person. In the U.S. however, under the first-to-invent regime, such a determination can involve protracted contested proceedings (called interferences) which take years to resolve. In addition, in the U.S., it is possible to contest the validity of a patent many years after its issuance (if the challenger can prove that someone else first invented the patented subject matter and did not withhold it from the public). Therefore, no matter how much information is published prior to issuance of a patent, unless the U.S. patent law changes from "first-to-invent" to "first-to-file", there will always be a risk of infringement that is incapable of assessment before the fact.

Despite its limitations, the new law marks an important step towards harmonization of the U.S. patent law with the laws of other industrialized nations. The hope is that the new law will make obtaining patents simpler, and reduce uncertainty, and therefore cost, in business.

Lastly, the new law does not appear to require many behavioral changes on the part of the patent applicants, but there are a few pitfalls to avoid. Examples are:

(i) Closer attention should be paid to the contents of a patent application. Its contents will be published and competitors can be sure that the published information is part of the U.S. application, a certainty absent under the old law.

(ii) Although the law permits portions unique to the U.S. application (absent from its published foreign counterparts) to be omitted from the U.S. publication document, if too much is omitted (so that the patent application as published does not comply with the requirements of U.S. law as to adequacy of disclosure) penalties can be imposed either by the USPTO or by the courts.

4 LIABILITY FOR I"GEMENT BEFORE PATENT GRANT

A] U.S. patent is a grant from the government to an inventor who creates a new, useful and unobvious process or product. This grant allows the inventor to prevent others from making, selling or using his/her invention for a limited time. This framework establishes that a patentee's rights arise from grant of the patent. Thus, there are no special rights associated with the patent application, and notably no right to exclude others. Indeed, under the old law, one could not sue for infringement of a U.S. patent application.

In exchange for requiring publication of patent applications, the patent law reform grants applicants for patents what has long been known in other countries as "provisional patent protection", i.e., the right to collect after issuance of the patent damages for infringement of a claim contained in the patent application as published. The damages accrue from the date of publication of the patent application but cannot actually be awarded until and unless the patent issues. The amount of damages that can be awarded is equal to a "reasonable royalty", i.e., the fee that a licensee of the patent would be willing to pay and that the licensor would be willing to accept in exchange for permitting the licensee to practice the patented invention.

A prerequisite to recovering damages is that the infringer must have had actual knowledge of the published application. This means that the infringer must have actually-in reality-seen the published patent application. The patent holder can receive compensation only for that portion of the pre-grant infringement during which the patent owner can show that the infringer had specific knowledge of the published patent application.

A second prerequisite to the recovery of damages for provisional infringement is that each infringed claim of the patent application must be identical or nearly identical to a claim of the issued patent. Similar provisions are contained in the patent laws of other countries. At a minimum, this means that the accused activity must infringe both the published and the granted claim. At first this appears to be a steep barrier because the claims of a patent application are almost always modified (narrowed to avoid prior references) before a patent is granted. Occasionally, broader claims are added which improve the scope of protection for the invention. The result is that the claims of the issued patent are typically substantially different from the original claims of the patent application.

The grant of provisional rights in the US. is intended to deter competitors from simply copying patent applications. The requirement that the infringed claim in the published patent application be "substantially identical" to a claim of the issued patent is designed to prevent patent applicants from claiming too broadly and thus attempting to deter competitors from pursuing activities they have every right to pursue.