ANSYS, Inc. v. Computational Dynamics North America, Ltd.

LYNCH, Chief Judge. ANSYS, Inc. (“ANSYS”) is a company that produces propriety software in the field of computational fluid dynamics (“CFD”). In August 2009, ANSYS sued its former employee, Dr. Doru Caraeni, and his new employer, Computational Dynamics North America, Ltd. (“CDNA”), in the federal district court for the district of New Hampshire, alleging breach by Dr. Caraeni of the noncompetition and confidentiality clauses in his ANSYS employment contract, interference with contractual relations, misappropriation of trade secrets, and unfair trade practices. CDNA and ANSYS are competitors; together they occupy approximately 80 percent of the market for CFD software, which is used for modeling and simulating fluid and gas flows. ANSYS had employed Dr. Caraeni from January 2002 to May 2009 to develop code for CFD simulation software. ANSYS sought a preliminary injunction to enforce the provisions of a one-year noncompetition clause in Dr. Caraeni’s employment agreement. The district court denied the request, finding that ANSYS had not met the requirements for injunctive relief. See ANSYS, Inc. v. Computational Dynamics N. Am., Ltd., Civ. No. 09-cv-284-SM, 2009 WL 4403745 (D.N.H. Nov. 25, 2009). We expedited the appeal. After oral argument this court issued an order affirming the denial of preliminary injunctive relief. We also said we would later issue our opinion in this case, as we do now, and judgment would not enter until we had done so. I. Our standard of review for denials of injunctive relief is strict: “we will reverse such a denial only if the district court mistook the law, clearly erred in its factual assessments, or otherwise abused its discretion in granting the preliminary injunction.” McClure v. Galvin, 386 F.3d *78 36, 41 (1st Cir.2004) (internal quotation marks omitted). While we think this case is close, we cannot say the court abused its discretion. When deciding whether to grant ANSYS preliminary injunctive relief, the district court was required to weigh four factors: (1) whether ANSYS had shown a likelihood of success on the merits, (2) whether ANSYS had shown that it would suffer irreparable harm if the injunction was denied, (3) the balance of the relevant hardships, and (4) any impact that the court’s ruling may have on the public interest. See Wine & Spirits Retailers, Inc. v. Rhode Island, 418 F.3d 36, 46 (1st Cir.2005). The first factor, likelihood of success, is usually given particularly heavy weight. Waldron v. George Weston Bakeries Inc., 570 F.3d 5, 8 (1st Cir.2009). The district court denied ANSYS’s motion on the grounds that the company had failed to demonstrate likely success on the merits and also failed to show that it would suffer irreparable harm absent injunctive relief. ANSYS, Inc., 2009 WL 4403745, at *5-7. We begin our review of the district court’s decision with the contract clause in question, which Dr. Caraeni had signed in September 2001. It reads, in relevant part, “I agree that for a period of one (1) year following termination of my employment with [ANSYS], I will not become an employee ... or in any way engage in or contribute my knowledge to a competitor of [ANSYS].” Perhaps recognizing how poorly drafted this clause is, ANSYS offers two pertinent narrowing constructions: (1) the clause only applies to those employees who have access to confidential information or trade secrets — thus, the clause would not apply to, for instance, a maintenance worker; and (2) the clause only applies to employees who are in a position to use that information at their new employer. This interpretation, of course, requires judicial reconstruction of the broadly drafted contract language. Against the backdrop of these constructions, ANSYS argues it has to show two things, and no more, to establish breach of contract: that Dr. Caraeni had access to confidential information while at ANSYS and that he is in a position to use this knowledge at CDNA. ANSYS asserts it met this burden on the evidence produced at the one-day evidentiary hearing before the district court. As a result, ANSYS argues, it has shown Dr. Caraeni was in breach of his contract and that it will succeed on the merits of its claims. Since likelihood of success on the merits normally weighs heavily in the preliminary injunction calculus, ANSYS urges this showing is sufficient to merit preliminary injunctive relief. ANSYS also argues that proving breach, and no more, was sufficient to demonstrate irreparable harm and further support its request for the preliminary injunction. ANSYS asserts that holding it to any greater showing on the likelihood of success or irreparable harm would effectively eliminate the noncompete clause from the contract. ANSYS says that for the non-compete clause to have any meaning, the clause must be read to give ANSYS some greater benefit than the protection it enjoys under common law for its trade secrets and confidential information. AN-SYS urges that the district court erred by analyzing this case as though it was simply a trade secrets case. After all, ANSYS notes, it did not seek a preliminary injunction on its trade secrets claim. ANSYS’s argument about the import of its contract clause is not irrational and one might well be sympathetic to it. See Cont'l Group, Inc. v. Kinsley, 422 F.Supp. *79 838, 843 (D.Conn.1976) (finding a covenant not to compete for eighteen months reasonable under New York law to the extent that it barred similar employment at a direct competitor working to develop an identical product). As Judge Newman observed in that case, it is unclear that the test for enforcing a noncompetition covenant should be the same as would be used in obtaining an injunction to bar disclosure of trade secrets or an injunction to enforce a covenant not to disclose trade secrets. The court explained, “[t]he non-competition covenant adds something to the protection available to the employer beyond what he would expect from the normal incidents of the employer-employee relationship or from a secrecy agreement.” Id. at 844. Here, the contract clause could be justified as a prophylaxis, protecting ANSYS in a situation that created a high risk of disclosure of its confidential information to a competitor. See id. at 845. Whether New Hampshire law would enforce an agreement so construed is quite a different matter. ANSYS’s arguments presume its noncompetition clause is enforceable. However, New Hampshire’s public policy discourages covenants not to compete, allowing them only “if the restraint is reasonable, given the particular circumstances of the case.” Merrimack Valley Wood Prods., Inc. v. Near, 152 N.H. 192, 876 A.2d 757, 762 (2005). No New Hampshire cases are directly on point, as the vast majority of noncompetition agreement cases considered by the state’s supreme court have turned on other issues, such as the protection of employers’ interest in customer good will. See, e.g., Concord Orthopaedics Prof'l Ass’n v. Forbes, 142 N.H. 440, 702 A.2d 1273, 1276 (1997); Technical Aid Corp. v. Allen, 134 N.H. 1, 591 A.2d 262, 266-67 (1991). 1 New Hampshire courts have, however, articulated general criteria by which we may assess ANSYS’s claims: To determine whether a restrictive covenant ancillary to an employment contract is reasonable, we engage in a three-part inquiry: first, whether the restriction is greater than necessary to protect the legitimate interests of the employer; second, whether the restriction imposes an undue hardship upon the employee; and third, whether the restriction is injurious to the public interest. ACAS Acquisitions (Precitech) Inc. v. Hobert, 155 N.H.381, 923 A.2d 1076, 1084 (2007). A covenant that fails any one of these criteria “is unreasonable and unenforceable.” Merrimack Valley Wood Prods., Inc., 876 A.2d at 762; see also id. at 763-64 (affirming that an agreement that broadly barred an employee from *80 working with any client who had transacted business with his employer in the past year was unenforceable, reasoning that it went “far beyond the defendant’s sphere of customer goodwill, and was more restrictive than necessary to protect the plaintiffs’ legitimate interests”). New Hampshire courts have further clarified that employers’ “legitimate interests” include “trade secrets that have been communicated to the employee during the course of employment” and “confidential information communicated by the employer to the employee, but not involving trade secrets.” Syncom Indus., Inc. v. Wood, 155 N.H. 73, 920 A.2d 1178, 1185 (2007). Whether ANSYS’s construction of the noncompete clause was reasonable— that is, whether it went no further than was necessary to protect its legitimate interests, imposed no undue hardship on Dr. Caraeni, and was not inconsistent with the public interest — all involve policy choices that have not been explicitly decided in this context by the New Hampshire courts or legislature. Absent such guidance, we cannot say ANSYS has clearly demonstrated its likelihood of success on the breach of contract claim, nor can we conclude that the district court abused its discretion by finding that ANSYS had failed to make this showing. We also cannot say that the district court abused its discretion when it found ANSYS had not shown the likelihood of irreparable injury. ANSYS’s claim that its showing of breach was sufficient to show irreparable harm again assumes that the covenant not to compete is enforceable. In any event, we reject ANSYS’s suggestion at oral argument that because of the difficulty of proving damages from breach of the covenant, as opposed to damages from the actual use of confidential material, we should deem its remedy at law inadequate. We further reject its related claim that this alleged inadequacy would justify equitable relief for breach of the covenant without any showing of injury. New Hampshire law has not adopted that approach. As to the district court’s analysis of AN-SYS’s injury, we will accept arguendo that Dr. Caraeni had access to confidential proprietary information. The district court concluded that even were that so, “Ansys has failed to demonstrate that [Dr. Caraeni] is likely to use that [confidential] information during the course of his employment at CDNA.” This conclusion was based on factual findings, which we cannot set aside unless they constituted clear error. E.g., McClure, 386 F.3d at 41. The district court explained that it found credible the testimony of Dr. Wayne Smith, who is the general manager of CDNA. In particular, the court accepted three aspects of Dr. Smith’s testimony: (1) “that CDNA maintains and enforces a strict policy preventing its employees from using confidential and trade secret information they may have acquired from prior employers;” (2) that Dr. Caraeni had not been assigned and, in the near term (that is, to the May 2010 date for expiration of the one-year period since his departure from ANSYS), will not be assigned “to perform any work at CDNA that might allow him to use any of ANSYS’s confidential or trade secret information;” and (3) that any trade secret or confidential information Dr. Caraeni had acquired at AN-SYS would not be useful to work on CDNA’s CFD software because of the different architecture of the two companies’ respective software codes. ANSYS, Inc., 2009 WL 4403745, at *6. ANSYS emphatically attacks the third of these findings in particular and urges *81 that it constitutes clear error. 2 Even if there were reason to be skeptical of the third finding — after all, CDNA chose to hire Dr. Caraeni for reasons that may have included what he learned in his prior employment — that would not make the finding clear error. More pertinently, the district court reasoned Dr. Caraeni would not perform any tasks that might require confidential information he had obtained at ANSYS. The court’s first two factual findings support its conclusion, and we cannot say that they were clearly erroneous. There is no need to discuss the myriad other arguments the parties have pressed. II. The district court’s denial of the preliminary injunction is affirmed. So ordered. Footnotes [1] Both parties note a single case in which the New Hampshire Supreme Court addressed a superficially similar attempt by an employer to enforce a covenant not to compete to prevent an employee from working for a competitor in a narrow field, upholding a district court’s determination that the noncompetition agreement at issue was enforceable. See ACAS Acquisitions (Precitech) Inc. v. Hobert, 155 N.H.381, 923 A.2d 1076, 1089 (2007). However, ACAS relied on a variety of facts not presented by this case. Id. at 1087-89. Moreover, to the limited extent that ANSYS relies on ACAS, it mischaracterizes the court's holding. The court did not, as ANSYS suggests, "presume[] irreparable harm from the presumed use of ... information.” Instead, it cited specific evidence "that the defendant intended to use his knowledge learned at [plaintiff's company] to help” his new employer, id. at 1083, 1085 (internal quotation marks omitted), and affirmed the district court's holding that the defendant’s “use and disclosure of ... information to or for the benefit of [his new employer] has caused, and will continue to cause,” the plaintiff harm, id. at 1087. [2] ANSYS also argues that the district court erred in crediting Dr. Smith's testimony, urging that his statements were self-serving and speculative. Given our traditional deference to district court credibility determinations, e.g., Jennings v. Jones, 587 F.3d 430, 444 (1st Cir.2009), we cannot say the court's decision was an abuse of discretion.