Western Electric Company, Inc. v. Stewart-Warner Corporation, Western Electric Company, Inc. v. Stewart-Warner Corporation

631 F.2d 333, 208 U.S.P.Q. (BNA) 183, 1980 U.S. App. LEXIS 13465

• Court: Court of Appeals for the Fourth Circuit
• Decided: October 2, 1980
• Docket: 79-1129, 79-1130
• Status: Published

Opinion

WIDENER, Circuit Judge:

Stewart-Warner Corp., defendant below, appeals from the judgment of the district court finding damages against it on account of patent infringement and deciding that its defenses of misuse and failure to comply with a consent antitrust decree in the district court in New Jersey were not well taken. Western Electric Co., plaintiff in the district court, appeals from the judgment of the district court that it not be awarded attorneys’ fees in this as an exceptional case and also increased damages for infringement. We affirm the judgment of the district court in all respects.

Western Electric brought this action for patent infringement, claiming that Stewart-Warner infringed its Derick-Frosch patent No. 2,802,760 issued on August 13, 1957. It is uncontested that Western Electric is the owner of this patent, and infringement is not now contested. Western Electric sought damages from Stewart- Warner for its infringement, and also sought that such damages be increased pursuant to 35 U.S.C. § 284 1 and that it be awarded attorneys’ fees pursuant to 35 U.S.C. § 285. 2

The trial below was bifurcated, with the first hearing dealing with validity only and the second with all other issues. Both stages were tried by a Special Master, with the district court adopting the Master’s, findings pursuant to FRCP 53(e)(2).

Prior to the first hearing, Stewart-Warner challenged the validity of the Derick-Frosch patent on numerous grounds and also claimed noninfringement. At the first hearing, however, Stewart-Warner abandoned its claim of invalidity based on anticipation under 35 U.S.C. § 102 and defended only on obviousness grounds. The first hearing resolved the question of obviousness in Western Electric’s favor. That finding is not challenged here. The second hearing resolved all other issues raised. At that stage, Stewart-Warner contended that the patent in suit, although valid, was unenforceable because of Western Electric’s misuse of that patent and violation of a court antitrust order in the district court in New Jersey. Both of those issues are argued on appeal. Various aspects of the award of damages were also contested in the second hearing, as is the case here.

The parties are involved in semiconductor manufacturing operations. The Derick-Frosch patent is a process patent that allows the use of silicon in the manufacture of semiconductors. It played an important role in the development of solid state devices and microelectronic components. Since the validity of the patent is not in issue and since a complete understanding of the intricacies of the process is not necessary for understanding the issues raised before us, we will not discuss the details of the patent nor its application in the manufacturing process. Following the briefing schedule, we will first deal with Western Electric’s claims of error and then discuss the errors raised by Stewart-Warner.

In deciding Western Electric’s claim for attorneys’ fees and increased damages, a brief discussion of the dealings between the parties is necessary. For approximately eleven years prior to the institution of this lawsuit, Western Electric and Stewart-Warner had negotiated and disputed over the Derick-Frosch patent. These protracted negotiations proved fruitless, with the parties never able to agree upon a licensing agreement for the patent in suit.

In 1964, negotiations between the parties regarding the licensing of the Derick-Frosch patent began. At that time Stewart-Warner was not manufacturing semiconductors which would have required the use of the Derick-Frosch patent. These negotiations continued at a snail’s pace. Western initially offered Stewart-Warner a license under one or all of its patents in its semiconductor portfolio. Later into the negotiations, Western indicated that if it took a royalty free grant-back license under Stewart-Warner’s portfolio of patents, it would reduce the royalty rate it charged to Stewart-Warner. Western would not, however, allow a royalty rate reduction based on a grant-back license arrangement if Stewart-Warner sought licensing only under the one Derick-Frosch patent.

Stewart-Warner instead sought a bilateral 3 license under the Derick-Frosch patent with a grant-back license and concomitant royalty reduction to Western. Such an arrangement was sought, says Stewart-Warner, because .it only needed a license under the one patent.

Western Electric would not agree to such an arrangement. It would only grant a unilateral license with a 1½% royalty rate for the Derick-Frosch patent alone. Western refused Stewart-Warner’s request, it claims, because it feared that in the future Stewart-Warner might claim that Western coerced it into such a one-sided agreement. Stewart-Warner sought to reassure Western by supplying a letter stating that the agreement was completely its idea and that' no coercion was involved, but Western was not so reassured.

Some time after the negotiations were initiated, counsel for Stewart-Warner began to investigate possible patent misuse and antitrust abuse by Western of the Derick-Frosch patent. The theory initially espoused was that such arose from Western’s licensing procedure in which it took grant-backs. Such procedure was discriminatory, counsel asserted, because it allowed Western to adjust the net balance royalty rate as it saw fit. Counsel reasoned that Western might have also violated a 1956 antitrust consent decree 4 which required Western to grant licenses for any, some or all of its patents for royalties that are non-discriminatory as between licensees. Counsel argued that refusal by Western to grant a royalty reduction through grant-back licensing when a potential licensee sought a license under one patent while allowing a royalty reduction if the licensee took a license under a package of patents was discriminatory and in violation of the consent decree. Such a violation, he asserted, would be tantamount to patent misuse. 5

The stalemate between the parties continued with Western seeking the portfolio cross-licensing package and Stewart-Warner seeking a bilateral license only on the one patent with grant-back licensing of its portfolio. The parties remained hopelessly deadlocked. Thereupon, Western filed this suit in 1975.

Western argues that the district court erred in not granting its request for increased damages and attorneys’ fees. These awards, which are statutorily established, are within the discretion of the trial court to award. Marvel Speciality Co. v. Bell Hosiery Mills, Inc., 386 F.2d 287 (4th Cir. 1967), cert. den. 390 U.S. 1030, 88 S.Ct. 1409, 20 L.Ed.2d 286 (1968); Colgate-Palmolive Co. v. Carter Products, Inc., 230 F.2d 855 (4th Cir.), cert. den. 352 U.S. 843, 77 S.Ct. 43, 1 L.Ed.2d 59 (1956). In fact, the statute permits the granting of attorneys’ fees only in exceptional cases. We cannot reverse the district court’s findings absent a finding that it abused its discretion in not making the awards that Western requested.

Western contends that circumstances do exist here to show that the trial court abused its discretion in not making the requested awards. It claims that Stewart-Warner knowingly and intentionally infringed the Derick-Frosch patent. It then asserts that Stewart-Warner deliberately delayed the licensing negotiations. Such a delay, it argues, shows bad faith on Stewart-Warner’s part so as to justify the awarding of increased damages and attorneys’ fees. The district court found that Stewart-Warner was “dilatory in its conduct of the license negotiations with Western.” Further, the court found that the license negotiations were one-sided in that Stewart-Warner refused to take a license except under its own terms. The court concluded, however, that Stewart-Warner had not acted in bad faith.

Western contends that this finding is clearly erroneous. It presents no arguments, however, that rebut the district court’s determination. We note, as did the district court, that the record shows that Western did not persist in its effort to enforce its patent in that it did not seek to litigate promptly the matter when it became apparent that the negotiations would not be successful. Instead, it chose to wait for years to bring this action. All of this being so, we cannot say that the district court abused its discretion in refusing to award increased damages because of Stewart-Warner’s negotiation practices.

Western next claims that increased damages are required because of Stewart-Warner’s unreasonable reliance on in-house counsel’s opinions regarding patent misuse by Western. That attorney was the chief negotiator for Stewart-Warner in its licensing negotiations with Western. During the process, he became concerned that the Derick-Frosch patent was unenforceable. He communicated this belief in at least three in-house memoranda, sent in 1966, 1969, and 1970. All three of these memo-randa were sent to D. A. Potter, Vice President of Stewart-Warner, with the 1966 memo also addressed to J. D. Coffey, General Manager of the Microcircuits Division. Copies of all three were sent to W. W. Miller, Vice President of Industrial Relations and Legal, and copies of the 1969 and 1970 memos were sent to Coffey and Bennett Archambault, President and Chairman of the Board of Stewart-Warner. Western first contends that there is no proof that Stewart-Warner’s decision makers relied upon the advice of counsel regarding the unenforceability of the Derick-Frosch patent. It argues that the only testimony in the record regarding such reliance is that of Coffey who was not a decision maker. That argument overlooks the fact, however, that copies of the three memoranda were sent to the decision makers of Stewart-Warner. It certainly is not incredible that these men read the opinion of counsel and then directed their actions accordingly. The finding of the master adopted by the district court that “there was sufficient evidence of reliance by Stewart-Warner on a supposed defense” is not in our opinion clearly erroneous.

Western further contends that even if there were reliance, that such reliance was unreasonable. Basing its contentions upon General Electric Co. v. Sciaky Brothers, Inc., 415 F.2d 1068 (6th Cir. 1969), Western claims that Stewart-Warner’s reliance on in-house counsel was unreasonable as a matter of law. Sciaky does not, however, require a finding that the reliance was unreasonable. It instead discussed the theory that liability for increased damages may not be available if there is an “honest doubt” as to liability. And the court stated that “[o]f course we recognize that a good-faith opinion by competent and independent patent counsel may be important evidence to be weighed on the issue of ‘honest doubt’ of patent validity.” Sciaky at 1073.

Indeed, in H. K. Porter Co., Inc. v. Goodyear Tire and Rubber Co., 536 F.2d 1115 (6th Cir. 1976), the same court which handed down Sciaky, construed that case in quite the opposite manner than that sought by Western. Goodyear in that case had opinions both from in-house counsel and outside counsel that its product would not infringe a valid patent. The court held “this evidence is competent to show that Goodyear had an ‘honest doubt’ as to the validity of the patents and as to non-infringement.” Porter at 1124. We think that Porter states the proper rule. Just because an attorney is in-house counsel does not mean that his opinions are necessarily suspect. While Stewart-Warner might be on firmer ground in this case had it sought the opinion of an outside attorney, it cannot be- said that the opinion of in-house counsel must simply be disregarded as a matter of law. The fact that the attorney was in-house counsel, of course, should be weighed along with all the other evidence in the case as to whether or not Stewart-Warner was acting in good faith or as Porter phrases the question whether it had an “honest doubt” as to liability. We reject the contention of Western that the in-house attorney’s opinions on patent misuse and whether or not Western had violated the New Jersey consent decree were absurd. Simply because the district court rejected these defenses does not justify an inference of absurdity. On the whole case, we agree with the conclusion of the master and the district court that “it cannot be said that it was an open and shut case all along for either party.”

Likewise, we find no error in the district court’s refusal to award attorneys’ fees. Such fees are usually awarded to prevent gross injustice and, as we have previously noted, are awarded only in exceptional cases. Stillman v. Edmund Scientific Co., 522 F.2d 798 (4th Cir. 1975); H. K. Porter Co. v. Black and Decker Mfg. Co., 518 F.2d 1177 (7th Cir. 1975). That is not the case here. The finding of the district court, approved by us, that this was not an open and shut case all along for either party, is sufficient to justify its refusal to award attorneys’ fees. There certainly was no gross injustice done to Western by Stewart-Warner’s defense of the claim.

In its cross-appeal, Stewart-Warner claims error in the district court’s finding that Western Electric did not misuse the Derick-Frosch patent.' It advances three grounds for such misuse. First, Western misused the patent by refusing to give Stewart-Warner a royalty reduction for a grant-back license of Stewart-Warner’s portfolio of patents in a license of the Derick-Frosch patent alone. Second, misuse is shown by Western’s insistence that Stewart-Warner pay royalties on the sales price of the finished semiconductor device instead of on the value of the chips themselves. Third, misuse is shown by Western’s refusal to settle this lawsuit únder the ⅛ and ½ liability formula Western had devised and used in the settlement of other such suits pending along with this one.

Stewart-Warner’s primary contention is that Western misused its patent by refusing to allow a grant-back reduction if Stewart-Warner took a license under the Derick-Frosch patent alone. The negotiations were initiated with Western offering a license under one or more of its patents with or without a grant-back reduction. There was discussion of the idea of licensing all or most of Western’s semiconductor patents. As the negotiations further continued, four of Western’s patents were set out for special consideration. 6 When Stewart-Warner decided it needed and wanted a license under the Derick-Frosch patent only, Western would not allow a grant-back reduction against a royalty of 1 ½%.

By refusing the grant-back, Stewart-Warner claims that Western committed patent misuse and especially that it violated a 1956 antitrust consent decree which Western had entered into which required that it grant licenses under “any, some or all” of its patents with a royalty which is “nondiscriminatory as between licensees.” United States v. Western Electric, Inc., No. 17-49 (D.N.J. Jan. 24, 1956). Stewart-Warner’s claim is that Western sought to coerce it into taking a package license of more than one product as opposed to a license under one patent only.

We think the above contention as to misuse by Stewart-Warner is not well taken. This is not a case in which Western refused to grant Stewart-Warner any license .other than a license of a package of patents, as in American Securit Co. v. Shatterproof Glass Corp., 268 F.2d 769 (3d Cir. 1959). Here, Western would not take a grant-back license so as to lower the cash royalties to be paid by Stewart-Warner. The fact that Stewart-Warner would have preferred to pay less cash, with the balance payable by a grant-back of a license to operate under its patents, is insufficient to make out a case of coercive package licensing. In order to prevail on this claim, Stewart-Warner must have shown that Western did not give it a choice to take a license under the Derick-Frosch patent alone or in combination with other patents on reasonable terms. Well Surveys, Inc. v. Perfo-Log, Inc., 396 F.2d 15 (10th Cir. 1968). Stewart-Warner has not made such a showing. It is undisputed that Western gave it the choice to take the Derick-Frosch patent alone at one rate of royalty or a combination of the Derick-Frosch patent with other patents at different royalty rates. As Well Surveys teaches, “freedom of choice is the controlling question,” at p. 18. This makes important the finding of the district court that “viable economic alternatives were made available to Stewart-Warner by Western through the application of its royalty reduction formula.” Stewart-Warner has not shown this finding to be clearly erroneous.

We also do not think Western was shown to have violated the 1956 decree of the district court in New Jersey. Stewart-Warner has not shown that it was treated any differently from any other potential licensee. The evidence showed that Western had never given a grant-back reduction when a single patent was licensed. In fact, the record shows that no one had ever sought the licensing arrangement Stewart-Warner proposed. It has not shown that Western granted the terms it sought to someone else and then refused to extend those same terms to Stewart-Warner. The 1956 decree does not require that all licensees be treated exactly the same, only that they not be discriminated against. The finding of no discrimination of the district court is amply supported by the record.

Stewart-Warner next claims patent misuse in Western Electric’s litigation settlement practices after the inception of and prior to the conclusion of this lawsuit. Stewart-Warner presents no facts in the record to show that the agreements reached in those instances were more than compromises of Western’s claims for royalties. We therefore agree with the district court that such action does not constitute patent misuse, and find no need to discuss Western’s various detailed defenses.

Stewart-Warner also claims patent misuse in Western’s method of computing royalties for the Derick-Frosch patent. Stewart-Warner claims that Western sought to extend its monopoly beyond the Derick-Frosch patent by basing royalties on the sales of the finished semiconductor devices instead of the fair market value of the chips. As the district court noted, such a method is simply more convenient because it is much easier to ascertain the selling price of the finished product than the fair market value of one of its components. Without more, such a provision is not patent misuse. A related subject with respect to damages is discussed further along in the opinion with supporting authorities.

Stewart-Warner’s reliance upon Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 89 S.Ct. 1562, 23 L.Ed.2d 129 (1969), is misplaced. Hazeltine’s licensing agreements required licensees to pay a royalty rate based on final sales of products that might not have used any of Hazeltine’s patents. The Court held that conditioning the granting of licenses upon royalty payment on products which do not use the patent licenses amounted to patent misuse. It was not the royalty based upon a finished product that concerned the Court, but, instead, the requirement that Zenith pay royalties on all of its products manufactured whether Hazeltine’s patents were used or not. Such is not the case here. All of the housed semiconductors at issue make use of the Derick-Frosch patent. In fact, Western’s standard package licensing provides for the payment of royalties only on those products which include an invention owned by Western.

Finally, Stewart-Warner claims that the district court erred in the formula that it used to compute damages to be paid by Stewart-Warner to Western Electric. The parties below stipulated that the single patent royalty rate for the Derick-Frosch patent was 1½%. The parties also stipulated the dollar amounts to be used as a base for calculating damages under three different theories (1) the dollar amount of the finished semiconductor devices transferred by Stewart-Warner within the period in question; (2) the dollar amount of the chips made by Stewart-Warner in the United States and transferred by it to offshore assembly plants, plus the dollar amount of finished semiconductors made completely in this country; and (3) the dollar amount of chips made by Stewart-Warner and transferred to offshore assembly facilities, plus the dollar amount of chips used in semiconductors processed in this country. The district court assessed damages based upon the finished semiconductors represented by the first theory above.

An understanding of the difference between a chip and the semiconductor becomes of some consequence now. Briefly, silicon is formed into cylinders and sliced into wafers. The Derick-Frosch process then comes into play scoring the wafers with tiny squares. The wafer is split along the squares with each resulting square, then called a chip, used in a semiconductor device. A semiconductor device consists of a chip with many leads connected to various parts of the chip. This chip with its leads are housed in plastic to hold the leads to the chip in place. The leads are then connected to the electronic device in which the semiconductor is used.

After thoroughly examining this relationship, the district court concluded that the market value of the housed conductor was the appropriate base for determining damages. It concluded that substantially all the market value of the device was derived from the chip itself, not the leads or the plastie enclosure. The process patented under the Derick-Frosch allowed the use of silicon to replace germanium, which had been less satisfactory, and as such the patent was a leading one in its field.

The court also noted that the base used by Western in its licensing contracts was the housed device and not the chip, and that if Western had in fact used the value of the chip, it probably would have used a substantially larger royalty rate than it was using. The court also considered the convenience of such a method because of the easily identifiable market value of the finished product and the fact that the chip itself is not a regularly marketed product.

Included in the total damages was the value of some housed devices in which the housings were made outside the United States. Stewart-Warner manufactured the chips within the United States and shipped them out of the country for encapsulation. The finished product was then shipped back to Stewart-Warner. Stewart-Warner claims that the court below committed error in basing the damage award upon the housed devices instead of the chips themselves because by so doing the court gave extraterritorial effect to the Derick-Frosch patent and misapplied the entire market value rule. Its first argument is easily disposed of.

We think Western’s patent monopoly is not being extended to production outside the United States contra to 35 U.S.C. § 271. 7 Stewart-Warner’s infringement occurred within the United States. Only the encapsulation occurred outside. In addition, the court did not conclude that Stewart-Warner sold the chips to an independent business outside the United States. Instead, it held, considering all of the facts, that Stewart-Warner merely shipped the chips to a bailee outside the United States who housed them and returned them to Stewart-Warner. Therefore, the court concluded that it was appropriate to include these devices in the final measure of damages. We agree with this conclusion.

In Marvel Specialty Co. v. Bell Hosiery Mill, Inc., 386 F.2d 287 (4th Cir. 1967), we recognized the well accepted rule that if a royalty has been established for the patented product, that royalty rate is the best measure of damages. Here, Western Electric had a well established licensing program under which it licensed the Derick-Frosch and other patents involved in the production of semiconductors based upon the selling price of the housed device. What Stewart-Warner wants us to do is to hold the 1½% rate intact, but significantly reduce the base. Under the case law, however, the district court can award damages based on an established royalty. 8 In fact, the statute provides at least an award of a reasonable royalty. 9 The action of the district court in providing for damages according to a previously established royalty rate is without error.

Stewart-Warner contends additionally that the district court incorrectly applied the entire market value rule to this case. The court below found that the value of the device was “primarily, if not entirely” derived from the chip itself. The fact that the cost of producing the chips is small as compared to the total cost of production of the final product is not necessarily determinative. A more critical fact is that the semiconductor derives substantially all, if not its entire, value from the use of the Derick-Frosch patent. The district court therefore correctly determined that the sales price of the finished product is a proper base for the computation of damages. In making such a determination, that court correctly applied the principle set out in Westinghouse Co. v. Wagner Mfg. Co., 225 U.S. 604, 614-15, 32 S.Ct. 691, 694, 56 L.Ed. 1222 (1912), where the Court stated that damages (referring there to profits) are properly calculated on the entire product when “the entire value of the whole machine, as a marketable article, is properly and legally attributable to the patented feature.” [Citation omitted.]

Accordingly, the judgment of the district court is affirmed. Each party to the appeal will bear its own costs.

AFFIRMED.

1 .35 U.S.C. § 284 provides in part:

Upon finding for the claimant the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer, together with interest and costs as fixed by the court.

When the damages are not found by a jury, the court shall assess them. In either event the court may increase the damages up to

three times the amount found or assessed

2 . 35 U.S.C. § 285 provides:

The court in exceptional cases may award reasonable attorney fees to the prevailing party.

3 . A unilateral license as referred to by the parties is a license by the patent owner to another to use the patent in return for a royalty. A bilateral license as so referred to is when each of the parties grants to the other the right to use certain patents belonging to the respective parties.

4 . United States v. Western Electric, Inc., No. 17 49 (D.N.J.1956).

5 . See American Securit Co. v. Shatterproof Glass Co., 268 F.2d 769 (3d Cir.), cert. den. 361 U.S. 902, 80 S.Ct. 210, 4 L.Ed.2d 157 (1959).

6 . Those patents were Nos. 3,122,817 (Andrus 4); 2,950,425 (Pfann 15); 3,165,811 (Kleim-ack 10); and 2,802,760 (Derick Frosch).

7 . 35 U.S.C. § 271 provides m pertinent part:

(a) Except as otherwise provided in this title, whoever without authority makes, uses or sells any patented invention, within the United States during the term of the patent therefor, infringes the patent. .

8 . Marvel Specialty Co. v. Bell Hosiery Mills, Inc., supra; Tektronix, Inc. v. United States, 552 F.2d 343 (Ct.Cl.1977).

9 . 35 U.S.C. § 284.