BDO Seidman v. Hirshberg

*387 OPINION OF THE COURT Levine, J. BDO Seidman (BDO), a general partnership of certified public accountants, appeals from the affirmance of an order of Supreme Court granting summary judgment dismissing its complaint against defendant, who was formerly employed as an accountant with the firm. The central issue before us is whether the “reimbursement clause” in an agreement between the parties, requiring defendant to compensate BDO for serving any client of the firm’s Buffalo office within 18 months after the termination of his employment, is an invalid and unenforceable restrictive covenant. The courts below so held. Facts and Procedural History BDO is a national accounting firm having 40 offices throughout the United States, including four in New York State. Defendant began employment in BDO’s Buffalo office in 1984, when the accounting firm he had been working for was merged into BDO, its partners becoming BDO partners. In 1989, defendant was promoted to the position of manager, apparently a step immediately below attaining partner status. As a condition of receiving the promotion, defendant was required to sign a “Manager’s Agreement,” the provisions of which are at issue. In paragraph “sixth” defendant expressly acknowledged that a fiduciary relationship existed between him and the firm by reason of his having received various disclosures which would give him an advantage in attracting BDO clients. Based upon that stated premise, defendant agreed that if, within 18 months following the termination of his employment, he served any former client of BDO’s Buffalo office, he would compensate BDO “for the loss and damages suffered” in an amount equal to IV2 times the fees BDO had charged that client over the last fiscal year of the client’s patronage. Defendant was to pay such amount in five annual installments. Defendant resigned from BDO in October 1993. This action was commenced in January 1995. During pretrial discovery, BDO submitted a list of 100 former clients of its Buffalo office, allegedly lost to defendant, who were billed a total of $138,000 *388 in the year defendant left the firm’s practice. Defendant denied serving some of the clients, averred that a substantial number of them were personal clients he had brought to the firm through his own outside contacts, and also claimed that with respect to some clients, he had not been the primary BDO representative servicing the account. Following discovery, the parties exchanged motions for summary judgment. BDO’s submissions on the motion did not contain any evidence that defendant actually solicited former clients, and did not rely in any way on claims that defendant used confidential information in acquiring BDO clients. Supreme Court granted summary judgment to defendant, concluding that the reimbursement clause was an overbroad and unenforceable anti-competitive agreement. The Appellate Division agreed, holding that the entire agreement was invalid (247 AD2d 923). Discussion Concededly, the Manager’s Agreement defendant signed does not prevent him from competing for new clients, nor does it expressly bar him from serving BDO clients. Instead, it requires him to pay “for the loss and damages” sustained by BDO in losing any of its clients to defendant within 18 months after his departure, an amount equivalent to IV2 times the last annual billing for any such client who became the client of defendant. Nonetheless, it is not seriously disputed that the agreement, in its purpose and effect, is a form of ancillary employee anti-competitive agreement that will be carefully scrutinized by the courts (see, Columbia Ribbon & Carbon Mfg. Co. v A-1-A Corp., 42 NY2d 496, 499). Reported cases adjudicating the validity of post-employment restrictive covenants go back almost 300 years (see, Blake, Employee Agreements Not To Compete, 73 Harv L Rev 625, 629, citing, inter alia, Mitchel v Reynolds, 1 P Wms 181, 24 Eng Rep 347 [QB 1711]). In the 19th century, a standard of reasonableness for judging the validity of such agreements developed in case law here and in England, balancing the need of fair protection for the benefit of the employer against the opposing interests of the former employee and the public (Blake, op. cit., at 639-640). The modern, prevailing common-law standard of reasonableness for employee agreements not to compete applies a three-pronged test. A restraint is reasonable only if it: (1) is no greater than is required for the protection of the legitimate interest of the employer, (2) does not impose undue hardship *389 on. the employee, and (3) is not injurious to the public (see, e.g., Technical Aid Corp. v Allen, 134 NH 1, 8, 591 A2d 262, 265-266; Blake, op. cit., at 648-649; Restatement [Second] of Contracts § 188). A violation of any prong renders the covenant invalid. New York has adopted this prevailing standard of reasonableness in determining the validity of employee agreements not to compete. “In this context a restrictive covenant will only be subject to specific enforcement to the extent that it is reasonable in time and area, necessary to protect the employer’s legitimate interests, not harmful to the general public and not unreasonably burdensome to the employee” (Reed, Roberts Assocs. v Strauman, 40 NY2d 303, 307). In general, we have strictly applied the rule to limit enforcement of broad restraints on competition. Thus, in Reed, Roberts Assocs. (supra), we limited the cognizable employer interests under the first prong of the common-law rule to the protection against misappropriation of the employer’s trade secrets or of confidential customer lists, or protection from competition by a former employee whose services are unique or extraordinary (40 NY2d, at 308). With agreements not to compete between professionals, however, we have given greater weight to the interests of the employer in restricting competition within a confined geographical area. In Gelder Med. Group v Webber (41 NY2d 680) and Karpinski v Ingrasci (28 NY2d 45), we enforced total restraints on competition, in limited rural locales, permanently in Karpinski and for five years in Gelder. The rationale for the differential application of the common-law rule of reasonableness expressed in our decisions was that professionals are deemed to provide “unique or extraordinary” services (see, Reed, Roberts Assocs. v Strauman, supra, 40 NY2d, at 308). BDO urges that accountancy is entitled to the status of a learned profession and, as such, the Karpinski and Gelder Medical Group precedents militate in favor of the validity of the restrictive covenant here. We agree that accountancy has all the earmarks of a learned profession. CPAs are required to have extensive formal training and education (Education Law § 7404 [1], [2], [3]; 8 NYCRR 70.1); they must pass a written examination (Education Law § 7404 [1], [4]; 8 NYCRR 70.3); and they are subject to mandatory continuing education requirements (Education Law § 7409; 8 NYCRR 70.6). Their professional conduct is regulated by the Board of Regents under *390 statutory disciplinary procedures (Education Law §§ 6509-6511). Moreover, there is a national code of professional conduct for certified public accountants which provides that “[hi]embers should accept the obligation to act in a way that will serve the public interest, honor the public trust, and demonstrate commitment to professionalism” (American Institute of Certified Public Accountants Code of Professional Conduct § 53, art II). The foregoing factors closely correspond to the criteria for a learned profession listed in Matter of Freeman (34 NY2d 1, 7). 1 Nonetheless, Gelder Medical Group and Karpinski do not dictate the result here. As we noted in Karpinski, the application of the test of reasonableness of employee restrictive covenants focuses on the particular facts and circumstances giving context to the agreement (28 NY2d, at 49; see also, Reed, Roberts Assocs. v Strauman, supra, 40 NY2d, at 307). This Court’s rationale for giving wider latitude to covenants between members of a learned profession because their services are unique or extraordinary (Reed, Roberts Assocs. v Strauman, supra) does not realistically apply to the actual context of the anti-competitive agreement here. In the instant case, BDO is a national accounting firm seeking to enforce the agreement within a market consisting of the entirety of a major metropolitan area. Moreover, defendant’s unchallenged averments indicate that his status in the firm was not based upon the uniqueness or extraordinary nature of the accounting services he generally performed on behalf of the firm, but in major part on his ability to attract a corporate clientele. Nor was there any proof that defendant possessed any unique or extraordinary ability as an accountant that would give him a competitive advantage over BDO. Moreover, the contexts of the agreements not to compete in Karpinski and Gelder Medical Group were entirely different. In each case, the former associate would *391 have been in direct competition with the promisee-practitioner for referrals from a narrow group of primary health providers in a rural geographical market for their medical or dental practice specialty. Thus, our learned profession precedents do not obviate the need for independent scrutiny of the anti-competitive provisions of the Manager’s Agreement under the tripartite common-law standard. Close analysis of paragraph sixth of the agreement under the first prong of the common-law rule, to identify the legitimate interest of BDO and determine whether the covenant is no more restrictive than is necessary to protect that interest, leads us to conclude that the covenant as written is overbroad in some respects. BDO claims that the legitimate interest it is entitled to protect is its entire client base, which it asserts a modern, large accounting firm expends considerable time and money building and maintaining. However, the only justification for imposing an employee agreement not to compete is to forestall unfair competition (see, Columbia Ribbon & Carbon Mfg. Co. v A-1-A Corp., 42 NY2d, supra, at 499). It seems self-evident that a former employee may be capable of fairly competing for an employer’s clients by refraining from use of unfair means to compete. If the employee abstains from unfair means in competing for those clients, the employer’s interest in preserving its client base against the competition of the former employer is no more legitimate and worthy of contractual protection than when it vies with unrelated competitors for those clients. Legal scholars and courts have more circumspectly identified the employer’s legitimate interest in employee anti-competitive agreements than that of preservation of the employer’s entire client base where, as here, there is no evidence that the employee obtained a competitive advantage by using confidential information. Professor Blake, in his seminal article in the Harvard Law Review, explains that the legitimate purpose of an employer in connection with employee restraints is “to prevent competitive use, for a time, of information or relationships which pertain peculiarly to the employer and which the employee acquired in the course of the employment” (Blake, op. cit., at 647 [emphasis supplied]). Protection of customer relationships the employee acquired in the course of employment may indeed be a legitimate interest (id., at 653). “The risk to the employer reaches a maximum in situations in which the employee must work closely with the client or customer over a long period of time, especially when his services are a *392 significant part of the total transaction” (id., at 661). Then, the employee has been enabled to share in the goodwill of a client or customer which the employer’s over-all efforts and expenditures created. The employer has a legitimate interest in preventing former employees from exploiting or appropriating the goodwill of a client or customer, which had been created and maintained at the employer’s expense, to the employer’s competitive detriment (see, Technical Aid Corp. v Allen, supra, 134 NH, at 9, 591 A2d, at 266; Peat Marwick Main & Co. v Haass, 818 SW2d 381, 387 [Tex Sup Ct]). It follows from the foregoing that EDO’s legitimate interest here is protection against defendant’s competitive use of client relationships which EDO enabled him to acquire through his performance of accounting services for the firm’s clientele during the course of his employment (Blake, op. cit., at 647-661). Extending the anti-competitive covenant to EDO’s clients with whom a relationship with defendant did not develop through assignments to perform direct, substantive accounting services would, therefore, violate the first prong of the common-law rule: it would constitute a restraint “greater than is needed to protect” these legitimate interests (Restatement [Second] of Contracts § 188 [1] [a]). 2 The foregoing overbreadth was the basis upon which a number of State courts have invalidated restrictive covenant agreements prohibiting employees from providing post-employment accounting services to any client of the employer without regard to whether the employee served the client during the course of employment (see, Peat Marwick Main & Co. v Haass, 818 SW2d 381, supra; Singer v Habif, Arogeti & Wynne, 250 Ga 376, 297 SE2d 473; Smith Batchelder & Rugg v Foster, 119 NH 679, 406 A2d 1310; Johnson & Co. v Salmen, 211 Neb 123, 317 NW2d 900). Although other courts have enforced employee restrictive covenants extending to an accounting firm’s entire client base, they have either failed adequately to identify the employer’s legitimate interest (see, Dobbins, DeGuire & Tucker u Rutherford, MacDonald & Olson, 218 Mont 392, 708 P2d 577; Follmer, Rudzewicz & Co. v Kosco, 420 Mich 394, 362 NW2d 676), or having properly identified the interest, did not engage in analysis as to whether the restriction was greater than necessary to protect it (see, Perry v Moran, 109 Wash 2d 691, 748 P2d 224, mod on reh 111 Wash 2d 885, 766 P2d 1096). *393 To the extent, then, that paragraph sixth of the Manager’s Agreement requires defendant to compensate EDO for lost patronage of clients with whom he never acquired a relationship through the direct provision of substantive accounting services during his employment, the covenant is invalid and unenforceable. By a parity of reasoning, it would be unreasonable to extend the covenant to personal clients of defendant who came to the firm solely to avail themselves of his services and only as a result of his own independent recruitment efforts, which BDO neither subsidized nor otherwise financially supported as part of a program of client development. Because the goodwill of those clients was not acquired through the expenditure of BDO’s resources, the firm has no legitimate interest in preventing defendant from competing for their patronage. Indeed, enforcement of the restrictive covenant as to defendant’s personal clients would permit BDO to appropriate goodwill created and maintained through defendant’s efforts, essentially turning on its head the principal justification to uphold any employee agreement not to compete based on protection of customer or client relationships. Except for the overbreadth in the foregoing two respects, the restrictions in paragraph sixth do not violate the tripartite common-law test for reasonableness. The restraint on serving EDO clients is limited to 18 months, and to clients of BDO’s Buffalo office. The time constraint appears to represent a reasonably brief interlude to enable the firm to replace the client relationship and goodwill defendant was permitted to acquire with some of its clients. Defendant is free to compete immediately for new business in any market and, if the over-broad provisions of the covenant are struck, to retain his personal clients and those clients of EDO’s that he had not served to any significant extent while employed at the firm. He has averred that BDO’s list of lost accounts contains a number of clients in both categories. Thus, there is scant evidence suggesting that the covenant, if cured of overbreadth, would work an undue hardship on defendant. Moreover, given the likely broad array of accounting services available in the greater Buffalo area, and the limited remaining class of BDO clientele affected by the covenant, it cannot be said that the restraint, as narrowed, would seriously impinge on the availability of accounting services in the Buffalo area from which the public may draw, or cause any significant dislocation in the market or create a monopoly in accounting services in that locale. These factors militate against *394 a conclusion that a reformed paragraph sixth would violate the third prong of the common-law test, injury to the public interest (see, Gelder Med. Group v Webber, supra, 41 NY2d, at 685; Blake, op. cit., at 686-687; Restatement [Second] of Contracts § 188, comment g). Severance or Partial Enforcement We conclude that the Appellate Division erred in holding that the entire covenant must be invalidated, and in declining partially to enforce the covenant to the extent necessary to protect BDO’s legitimate interest. The Appellate Division rejected partial enforcement or severance of the invalid part of the covenant, because “the court would thereby be required to rewrite the entire covenant” (247 AD2d, at 923). In Karpinski v Ingrasci (28 NY2d, supra, at 51-52), this Court expressly recognized and applied the judicial power to sever and grant partial enforcement for an overbroad employee restrictive covenant. The Court refused to give effect to the portion of the covenant which barred the practice of general dentistry, but enforced it respecting the practice of oral surgery, that being the employer’s actual, specialized dental practice. The issue of whether a court should cure the unreasonable aspect of an overbroad employee restrictive covenant through the means of partial enforcement or severance has been the subject of some debate among courts and commentators (see, Blake, op. cit, at 682-683). A legitimate consideration against the exercise of this power is the fear that employers will use their superior bargaining position to impose unreasonable anti-competitive restrictions, uninhibited by the risk that a court will void the entire agreement, leaving the employee free of any restraint (id.). The prevailing, modern view rejects a per se rule that invalidates entirely any overbroad employee agreement not to compete. Instead, when, as here, the unenforceable portion is not an essential part of the agreed exchange, a court should conduct a case specific analysis, focusing on the conduct of the employer in imposing the terms of the agreement (see, Restatement [Second] of Contracts § 184). Under this approach, if the employer demonstrates an absence of overreaching, coercive use of dominant bargaining power, or other anti-competitive misconduct, but has in good faith sought to protect a legitimate business interest, consistent with reasonable standards of fair dealing, partial enforcement may be justified (see, Blake, op. cit, at 633; Restatement [Second] of Contracts § 184 [1], [2]). We essentially adopted this more flexible position in Karpinski (supra). *395 Here, the undisputed facts and circumstances militate in favor of partial enforcement. The covenant was not imposed as a condition of defendant’s initial employment, or even his continued employment, but in connection with promotion to a position of responsibility and trust just one step below admittance to the partnership. There is no evidence of coercion or that the Manager’s Agreement was part of some general plan to forestall competition. Moreover, no proof was submitted that EDO imposed the covenant in bad faith, knowing full well that it was overbroad. Indeed, as already discussed, the existence of our “learned profession” precedents, and decisions in other States upholding the full terms of this type of agreement, support the contrary conclusion. Therefore, partial enforcement of paragraph sixth is warranted. The Appellate Division’s fear that partial enforcement will require rewriting the parties’ agreement is unfounded. No additional substantive terms are required. The time and geographical limitations on the covenant remain intact. The only change is to narrow the class of BDO clients to which the covenant applies (cf.,Karpinski v Ingrasci, supra [narrowing the scope of the prohibitive post-employment activity]). Moreover, to reject partial enforcement based solely on the extent of necessary revision of the contract resembles the now-discredited doctrine that invalidation of an entire restrictive covenant is required unless the invalid portion was so divisible that it could be mechanically severed, as with a “judicial blue pencil” (see, Blake, op. cit., at 681). The Restatement (Second) of Contracts rejected that rigid requirement of strict divisibility before a covenant could be partially enforced (see, Reporter’s Note, Restatement [Second] of Contracts § 184, at 32). Thus, we conclude that severance is appropriate, rendering the restrictive covenant partially enforceable. Damages Since defendant does not dispute that at least some EDO clients to which the restrictive covenant validly applies were served by him during the contractual duration of the restraint, plaintiff is entitled to partial summary judgment on the issue of liability. Remittal is required in order to establish plaintiffs damages, including resolution of any contested issue as to which of EDO’s former clients served by defendant the restrictive covenant validly covers. As to those clients, the measure of plaintiffs damages will depend in the first instance on the validity of the clause in *396 paragraph sixth of the Manager’s Agreement requiring defendant to compensate BDO “for the loss and damages suffered” in an amount equal to 1½ times the fees charged each lost client over the last full year the client was served by the firm. This provision essentially represents a liquidated damages clause, as BDO conceded at nisi prius. Liquidated damages provisions, under our precedents, are valid if the “damages flowing from a breach are difficult to ascertain [and under] a provision fixing the damages in advance * * * the amount is a reasonable measure of the anticipated probable harm” (City of Rye v Public Serv. Mut. Ins. Co., 34 NY2d 470, 473). On the other hand, if “the amount fixed is plainly or grossly disproportionate to the probable loss, the provision calls for a penalty and will not be enforced” (Truck Rent-A-Ctr. v Puritan Farms 2nd, 41 NY2d 420, 425). The damages here are sufficiently difficult to ascertain to satisfy the first requirement of a valid liquidated damages provision. Because of the inability to project with any degree of certainty how long a given client would have remained with EDO if defendant had not made himself available as an alternative source of accounting services, EDO’s actual lost profits from defendant’s breach would be impossible to determine with any precision. EDO submits that the second requirement is also met, in that the agreed upon amount is not grossly disproportionate to the anticipated probable harm. In rationalizing this measure of damages, EDO relies upon the affidavit of the managing partner of its Buffalo office that “the amount of liquidated damages was tied to a commonly accepted way of valuing a client account of a professional services firm as an asset” (emphasis supplied). It is apparent that BDO’s rationale for the liquidated damages formula likens the value of BDO’s loss of a client account to what an arm’s length purchaser would have paid for it as a firm asset on- a sale of the practice. Various courts reviewing accountants’ employee agreements not to compete have been persuaded of the reasonableness of a liquidated damages formula based on the prevalent use of client gross billings to value an accounting practice for purpose of acquisition (see, Holloway v Faw, Casson & Co., 319 Md 324, 355, 572 A2d 510, 525; Foti v Cook, 220 Va 800, 807, 263 SE2d 430, 434; Knight, Vale & Gregory v McDaniel, 37 Wash App 366, 372, 680 P2d 448, 453). In our view, however, the averment regarding the basis of the liquidated damages formula by no means conclusively *397 demonstrates the absence of gross disproportionality. Indeed, the nonspecific averment in the affidavit, quoted above, which EDO relies upon, is the only record evidence supporting the reasonableness of the liquidated damages clause. We note that other courts have remitted on the issue of the validity of these types of liquidated damages provisions in accountant employee anti-competitive agreements when they found the record insufficiently developed to establish that the amount fixed in the agreement was not so excessive to actual damages as to constitute a penalty (see, Perry v Moran, supra, 111 Wash, at 887, 766 P2d, at 1097; Follmer, Rudzewicz & Co. v Kosco, supra, 420 Mich, at 409, 362 NW2d, at 684). The sparse proof on this issue here persuades us that we, similarly, should remit for further development of the record on the liquidated damages formula. Accordingly, the order of the Appellate Division, insofar as appealed from, should be modified, without costs, by denying defendant’s motion for summary judgment, granting plaintiffs motion for partial summary judgment declaring the restrictive covenant enforceable as here provided, and remitting to Supreme Court for further proceedings in accordance with this opinion and, as so modified, affirmed. Chief Judge Kaye and Judges Bellacosa, Smith, Ciparick, Wesley and Rosenblatt concur. Order, insofar as appealed from, modified, without costs, by denying defendant’s motion for summary judgment, granting plaintiffs motion for partial summary judgment declaring the restrictive covenant enforceable as provided in the opinion, and remitting to Supreme Court for further proceedings in accordance with the opinion herein and, as so modified, affirmed. Footnotes [1] Law firm partnership agreements represent an exception to the liberality with which we have previously treated restraints on competition in the learned professions (see, Cohen v Lord, Day & Lord, 75 NY2d 95; Denburg v Parker Chapin Flattau & Klimpl, 82 NY2d 375). Our decisions invalidating anti-competitive clauses in such agreements were not based on application of the common-law rule, but upon enforcement of the public policy reflected in DR 2-108 (A) of the Code of Professional Responsibility (see, 22 NYCRR 1200.13). There is no counterpart to DR 2-108 (A) in the rules regulating the ethical conduct of accountants. Hence, Cohen and Denburg do not require invalidation of the restrictive covenant here (see, Faw, Casson & Co. v Cranston, 375 A2d 463, 468 [Del Ch]; Mailman, Ross, Toyes & Shapiro v Edelson, 183 NJ Super 434, 440, 444 A2d 75, 78). [2] A different result might obtain had BDO submitted any proof that defendant had used confidential firm information to attract BDO clients with whom he had not had a relationship while employed there.