Foley v. Interactive Data Corp. (1988)

Annotate this Case
[L.A. No. 32148. Supreme Court of California. December 29, 1988.]

DANIEL D. FOLEY, Plaintiff and Appellant, v. INTERACTIVE DATA CORPORATION, Defendant and Respondent

(Opinion By Lucas, C. J., with Panelli, Arguelles, and Eagleson, JJ., concurring. Separate concurring and dissenting opinions by Broussard and Kaufman, JJ. Separate dissenting opinion by Mosk, J.)

COUNSEL

Gilbert & Sackman, Gilbert, Cooke & Sackman, Steven J. Kaplan and Robert W. Gilbert for Plaintiff and Appellant. [47 Cal. 3d 662]

Lynne G. McGinnis, Joseph Posner, Kenneth J. Sargoy, Fred M. Blum, Sargoy & Blum, Paul Hoffman, Gary Williams, Wylie A. Aitken, Victoria De Goff, Douglas de Vries, John Gardenal, John R. Hillsman, Ian Herzog, Peter Hinton, Harvey R. Levine, Leonard Sacks, Robert Steinberg, John C. McCarthy and Cliff Palefsky as Amici Curiae on behalf of Plaintiff and Appellant.

Proskauer, Rose, Goetz & Mendelsohn, Robert V. Kuenzel, Steven G. Drapkin, Jeffrey A. Berman and Harold M. Brody for Defendant and Respondent.

Latham & Watkins, Josel E. Krischer, Deanna P. George, Michael J. Breining, Paul, Hastings, Janofsky & Walker, Paul Grossman, Paul W. Crane, Jr., Mary Craig Calkins and Michele M. Dosoer as Amici Curiae on behalf of Defendant and Respondent.

OPINION

LUCAS, C. J.

After Interactive Data Corporation (defendant) fired plaintiff Daniel D. Foley, an executive employee, he filed this action seeking compensatory and punitive damages for wrongful discharge. In his second amended complaint, plaintiff asserted three distinct theories: (1) a tort cause of action alleging a discharge in violation of public policy (Tameny v. Atlantic Richfield Co. (1980) 27 Cal. 3d 167 [164 Cal. Rptr. 839, 610 P.2d 1330, 9 A.L.R.4th 314]), (2) a contract cause of action for breach of an implied-in-fact promise to discharge for good cause only (e.g., Pugh v. See's Candies, Inc. (1981) 116 Cal. App. 3d 311 [171 Cal. Rptr. 917] [all references are to this case rather than the 1988 posttrial decision appearing at 203 Cal.App.3d 743]), and (3) a cause of action alleging a tortious breach of the implied covenant of good faith and fair dealing (e.g., Cleary v. American Airlines, Inc. (1980) 111 Cal. App. 3d 443 [168 Cal.Rptr. 722]). The trial court sustained a demurrer without leave to amend, and entered judgment for defendant.

The Court of Appeal affirmed on the grounds (1) plaintiff alleged no statutorily based breach of public policy sufficient to state a cause of action pursuant to Tameny; (2) plaintiff's claim for breach of the covenant to discharge only for good cause was barred by the statute of frauds; and (3) plaintiff's cause of action based on breach of the covenant of good faith and fair dealing failed because it did not allege necessary longevity of employment or express formal procedures for termination of employees. We granted review to consider each of the Court of Appeal's conclusions. [47 Cal. 3d 663]

We will hold that the Court of Appeal properly found that plaintiff's particular Tameny cause of action could not proceed; plaintiff failed to allege facts showing a violation of a fundamental public policy. We will also conclude, however, that plaintiff has sufficiently alleged a breach of an "oral" or "implied-in-fact" contract, and that the statute of frauds does not bar his claim so that he may pursue his action in this regard. Finally, we will hold that the covenant of good faith and fair dealing applies to employment contracts and that breach of the covenant may give rise to contract but not tort damages.

Facts

[1] Because this appeal arose from a judgment entered after the trial court sustained defendant's demurrer, "we must, under established principles, assume the truth of all properly pleaded material allegations of the complaint in evaluating the validity" of the decision below. (Tameny v. Atlantic Richfield Co., supra, 27 Cal. 3d 167, 170; Alcorn v. Anbro Engineering, Inc. (1970) 2 Cal. 3d 493, 496 [86 Cal. Rptr. 88, 468 P.2d 216].)

According to the complaint, plaintiff is a former employee of defendant, a wholly owned subsidiary of Chase Manhattan Bank that markets computer-based decision-support services. Defendant hired plaintiff in June 1976 as an assistant product manager at a starting salary of $18,500. As a condition of employment defendant required plaintiff to sign a "Confidential and Proprietary Information Agreement" whereby he promised not to engage in certain competition with defendant for one year after the termination of his employment for any reason. The agreement also contained a "Disclosure and Assignment of Information" provision that obliged plaintiff to disclose to defendant all computer-related information known to him, including any innovations, inventions or developments pertaining to the computer field for a period of one year following his termination. Finally, the agreement imposed on plaintiff a continuing obligation to assign to defendant all rights to his computer-related inventions or innovations for one year following termination. It did not state any limitation on the grounds for which plaintiff's employment could be terminated.

Over the next six years and nine months, plaintiff received a steady series of salary increases, promotions, bonuses, awards and superior performance evaluations. In 1979 defendant named him consultant manager of the year and in 1981 promoted him to branch manager of its Los Angeles office. His annual salary rose to $56,164 and he received an additional $6,762 merit bonus two days before his discharge in March 1983. He alleges defendant's officers made repeated oral assurances of job security so long as his performance remained adequate. [47 Cal. 3d 664]

Plaintiff also alleged that during his employment, defendant maintained written "Termination Guidelines" that set forth express grounds for discharge and a mandatory seven-step pretermination procedure. Plaintiff understood that these guidelines applied not only to employees under plaintiff's supervision, but to him as well. On the basis of these representations, plaintiff alleged that he reasonably believed defendant would not discharge him except for good cause, and therefore he refrained from accepting or pursuing other job opportunities.

The event that led to plaintiff's discharge was a private conversation in January 1983 with his former supervisor, Vice President Richard Earnest. During the previous year defendant had hired Robert Kuhne and subsequently named Kuhne to replace Earnest as plaintiff's immediate supervisor. Plaintiff learned that Kuhne was currently under investigation by the Federal Bureau of Investigation for embezzlement from his former employer, Bank of America. fn. 1 Plaintiff reported what he knew about Kuhne to Earnest, because he was "worried about working for Kuhne and having him in a supervisory position ..., in view of Kuhne's suspected criminal conduct." Plaintiff asserted he "made this disclosure in the interest and for the benefit of his employer," allegedly because he believed that because defendant and its parent do business with the financial community on a confidential basis, the company would have a legitimate interest in knowing about a high executive's alleged prior criminal conduct.

In response, Earnest allegedly told plaintiff not to discuss "rumors" and to "forget what he heard" about Kuhne's past. In early March, Kuhne informed plaintiff that defendant had decided to replace him for "performance reasons" and that he could transfer to a position in another division in Waltham, Massachusetts. Plaintiff was told that if he did not accept a transfer, he might be demoted but not fired. One week later, in Waltham, Earnest informed plaintiff he was not doing a good job, and six days later, he notified plaintiff he could continue as branch manager if he "agreed to go on a 'performance plan.' Plaintiff asserts he agreed to consider such an arrangement." The next day, when Kuhne met with plaintiff, purportedly to present him with a written "performance plan" proposal, Kuhne instead informed plaintiff he had the choice of resigning or being fired. Kuhne offered neither a performance plan nor an option to transfer to another position. fn. 2 [47 Cal. 3d 665]

Defendant demurred to all three causes of action. After plaintiff filed two amended pleadings, the trial court sustained defendant's demurrer without leave to amend and dismissed all three causes of action. The Court of Appeal affirmed the dismissal as to all three counts. We will explore each claim in turn.

I. Tortious Discharge in Contravention of Public Policy

We turn first to plaintiff's cause of action alleging he was discharged in violation of public policy. Labor Code section 2922 provides in relevant part, "An employment, having no specified term, may be terminated at the will of either party on notice to the other. ..." [2] This presumption may be superseded by a contract, express or implied, limiting the employer's right to discharge the employee. (Strauss v. A. L. Randall Co. (1983) 144 Cal. App. 3d 514, 517 [194 Cal. Rptr. 520]; Drzewiecki v. H & R Block, Inc. (1972) 24 Cal. App. 3d 695, 703 [101 Cal. Rptr. 169]; see also cases cited in part II(B) of this opinion, post, p. 680 et seq.) Absent any contract, however, the employment is "at will," and the employee can be fired with or without good cause. fn. 3 But the employer's right to discharge an "at will" employee is still subject to limits imposed by public policy, since otherwise the threat of discharge could be used to coerce employees into committing crimes, concealing wrongdoing, or taking other action harmful to the public weal. fn. 4

Petermann v. International Brotherhood of Teamsters (1959) 174 Cal. App. 2d 184 [344 P.2d 25], first stated the foregoing principle. There, [47 Cal. 3d 666] the plaintiff, a union business agent, alleged he was discharged when he refused to testify falsely to a state legislative committee. The trial court granted judgment on the pleadings to defendant. The Court of Appeal found the plaintiff was an employee-at-will (see Lab. Code, § 2922) but noted that "the right to discharge an employee under such a contract may be limited by statute [citations] or by considerations of public policy." (174 Cal.App.2d at p. 188.) Overruling the trial court, the Court of Appeal declared: "It would be obnoxious to the interests of the state and contrary to public policy and sound morality to allow an employer to discharge any employee, whether the employment be for a designated or unspecified duration, on the ground that the employee declined to commit perjury, an act specifically enjoined by statute." (Id., at pp. 188-189.) fn. 5

[3] Similarly, Tameny v. Atlantic Richfield Co., supra, 27 Cal. 3d 167, 178, declared that a tort action for wrongful discharge may lie if the employer "condition[s] employment upon required participation in unlawful conduct by the employee." In Tameny, the plaintiff alleged he was fired for refusing to engage in price fixing in violation of the Cartwright Act and the Sherman Antitrust Act. (Id., at p. 170.) We held the trial court erred in sustaining Atlantic Richfield's demurrer to plaintiff's tort action for wrongful discharge. Writing for the majority, Justice Tobriner concluded that "an employer's authority over its employee does not include the right to demand that the employee commit a criminal act to further its interests. ... An employer engaging in such conduct violates a basic duty imposed by law upon all employers, and thus an employee who has suffered damages as a result of such discharge may maintain a tort action for wrongful discharge against the employer." (Id., at p. 178.)

Tameny arose from facts quite similar to Petermann; in both cases, an employee was discharged for his refusal to violate a penal statute. The plaintiff in Petermann, however, had framed his complaint in contract, and sought only back wages; the Tameny plaintiff sought tort damages. In upholding the claim in Tameny, we explained that the cause of action was not dependent on an express or implied promise in the employment contract, "but rather reflects a duty imposed by law upon all employers in order to implement the fundamental public policies embodied in the state's penal statutes." (27 Cal.3d at p. 176.) We noted also that the existence of a contractual relationship would not bar an injured party from seeking relief [47 Cal. 3d 667] through tort remedies when the "employer's discharge of an employee contravenes the dictates of public policy." (Id., at pp. 175, 177.) fn. 6

Summarizing this authority, the Court of Appeal in Koehrer v. Superior Court (1986) 181 Cal. App. 3d 1155 [226 Cal. Rptr. 820], stated at page 1166: "As Tameny explained, the theoretical reason for labeling the discharge wrongful in such cases is not based on the terms and conditions of the contract, but rather arises out of a duty implied in law on the part of the employer to conduct its affairs in compliance with public policy .... [T]here is no logical basis to distinguish in cases of wrongful termination for reasons violative of fundamental principles of public policy between situations in which the employee is an at-will employee and [those] in which the employee has a contract for a specified term. [4] The tort is independent of the term of employment." fn. 7

Other state courts have on occasion failed to draw the distinction between contract-based causes of action and those based on policies extrinsic to the terms of the agreement. For example, the Wisconsin Supreme Court adopted a "narrowly circumscribed public policy exception" to the employment-at-will doctrine. Nonetheless, it relied on the fact that the remedies usually available for wrongful discharge under the relevant statutes were limited to contractual remedies, such as reinstatement and backpay, to conclude that such contractual remedies "are the most appropriate remedies for public policy exception wrongful discharges since the primary concern in these actions is to make the wronged employee 'whole.'" (Brockmeyer v. Dun & Bradstreet (1983) 113 Wis.2d 561 [335 N.W.2d 834, 840-841].)

This characterization of the nature of the action contrasts with our Tameny analysis, in which we deemed the public-policy-based cause of action as "ex delicto," or arising "from a breach of duty growing out of the [47 Cal. 3d 668] contract," rather than "from a breach of a promise set forth in the contract" or "ex contractu." (27 Cal.3d at p. 175.) As we explained, "an employer's obligation to refrain from discharging an employee who refuses to commit a criminal act does not depend upon any express or implied '"promise[s] set forth in the [employment] contract"' [citation], but rather reflects a duty imposed by law upon all employers in order to implement the fundamental public policies embodied in the state's penal statutes. As such, a wrongful discharge suit exhibits the classic elements of a tort cause of action." (Id., at p. 176.) Thus, the Wisconsin court focused on contract remedies on the assumption that the underlying interest was to compensate the employee, whereas California cases have focused on the general social policies being advanced by recognition of the public-policy-based cause of action.

In Tameny, because there was no statute specifically barring an employer from terminating an employee who refused to act illegally, the court was required to consider whether, without the authority of an express prohibition on the reasons for discharge, the plaintiff's action could proceed. We concluded that "even in the absence of an explicit statutory provision prohibiting the discharge of a worker on such grounds, fundamental principles of public policy and adherence to the objectives underlying the state's penal statutes require the recognition of a rule barring an employer from discharging an employee who has simply complied with his legal duty and has refused to commit an illegal act." (27 Cal.3d at p. 174, fn. omitted.) The public policy to which we looked thus was one about which reasonable persons can have little disagreement, and which was "firmly established" at the time of discharge. fn. 8

The employee in Tameny claimed his termination was based on his refusal to engage in statutorily forbidden conduct at his employer's behest. We mentioned generally that an employer's ability to discharge at will "'may be limited by statute ... or by considerations of public policy.'" (27 Cal.3d at p. 172.) Several subsequent Court of Appeal cases have limited our holding to policies derived from statute. (See Shapiro v. Wells Fargo Realty Advisors (1984) 152 Cal. App. 3d 467, 477 [199 Cal. Rptr. 613]; Gray v. Superior Court (1986) 181 Cal. App. 3d 813, 819 [226 Cal. Rptr. 570]; Tyco Industries, Inc. v. Superior Court (1985) 164 Cal. App. 3d 148, 159 [211 Cal. Rptr. 540].) The Court of Appeal in the present case asserted, "[t]o successfully plead a cause of action under the [Tameny] theory, plaintiff must allege that he was terminated in retaliation for asserting his statutory rights, or for his refusal [47 Cal. 3d 669] to perform an illegal act at the request of the employer, or that his employer directly violated a statute by dismissing him."

At least three other Court of Appeal decisions addressing the issue of where policy giving rise to an action may be found have concluded in dicta that public policy, as a basis for a wrongful discharge action, need not be policy rooted in a statute or constitutional provision. (See Koehrer v. Superior Court, supra, 181 Cal. App. 3d 1155, 1165, 1167; Garcia v. Rockwell Internat. Corp. (1986) 187 Cal. App. 3d 1556, 1561 [232 Cal. Rptr. 490]; Dabbs v. Cardiopulmonary Management Services (1987) 188 Cal. App. 3d 1437, 1443-1444 [234 Cal. Rptr. 129].)

We do not decide in this case whether a tort action alleging a breach of public policy under Tameny may be based only on policies derived from a statute or constitutional provision or whether nonlegislative sources may provide the basis for such a claim. [5] Even where, as here, a statutory touchstone has been asserted, we must still inquire whether the discharge is against public policy and affects a duty which inures to the benefit of the public at large rather than to a particular employer or employee. For example, many statutes simply regulate conduct between private individuals, or impose requirements whose fulfillment does not implicate fundamental public policy concerns. Regardless of whether the existence of a statutory or constitutional link is required under Tameny, disparagement of a basic public policy must be alleged, and we turn now to determining whether plaintiff has done so here.

[6a] In the present case, plaintiff alleges that defendant discharged him in "sharp derogation" of a substantial public policy that imposes a legal duty on employees to report relevant business information to management. An employee is an agent, and as such "is required to disclose to [his] principal all information he has relevant to the subject matter of the agency." (2 Witkin, Summary of Cal. Law (9th ed. 1987) Agency & Employment, § 41, p. 53; see Loughlin v. Idora Realty Co. (1968) 259 Cal. App. 2d 619, 629 [66 Cal. Rptr. 747]; Jolton v. Minster Graf & Co. (1942) 53 Cal. App. 2d 516, 522 [128 P.2d 101].) [7] [6b] Thus, plaintiff asserts, if he discovered information that might lead his employer to conclude that an employee was an embezzler, and should not be retained, plaintiff had a duty to communicate that information to his principal. fn. 9 [47 Cal. 3d 670]

It is unclear whether the alleged duty is one founded in statute. No enactment expressly requires an employee to report relevant information concerning other employees to his employer, and none prohibits discharge of the employee for so doing. fn. 10 The 1872 Civil Code, however, attempted to codify the common law of master-servant relations; its provisions, now in the Labor Code, provide that "[o]ne who, for a good consideration, agrees to serve another, shall perform the service, and shall use ordinary care and diligence therein, so long as he is thus employed." (Lab. Code, § 2854.) It is not clear whether the duty to communicate relevant information is subsumed under the statutory duty of ordinary care, or is a separate duty not codified by the 1872 Legislature.

Whether or not there is a statutory duty requiring an employee to report information relevant to his employer's interest, we do not find a substantial public policy prohibiting an employer from discharging an employee for performing that duty. fn. 11 Past decisions recognizing a tort action for discharge in violation of public policy seek to protect the public, by protecting the employee who refuses to commit a crime (Tameny, supra, 27 Cal. 3d 167; Petermann, supra, 174 Cal.App.2d 184), who reports criminal activity to proper authorities (Garibaldi v. Lucky Food Stores, Inc. (9th Cir. 1984) 726 F.2d 1367, 1374; Palmateer v. International Harvester Co., supra, 421 N.E.2d 876, 879-880), or who discloses other illegal, unethical, or unsafe practices (Hentzel v. Singer Co. (1982) 138 Cal. App. 3d 290 [188 Cal. Rptr. 159, 35 A.L.R.4th 1015] [working conditions hazardous to employees]). No equivalent public interest bars the discharge of the present plaintiff. fn. 12 When [47 Cal. 3d 671] the duty of an employee to disclose information to his employer serves only the private interest of the employer, the rationale underlying the Tameny cause of action is not implicated. fn. 13

We conclude that the Court of Appeal properly upheld the trial court's ruling sustaining the demurrer without leave to amend to plaintiff's first cause of action.

II. Breach of Employment Contract

Plaintiff's second cause of action alleged that over the course of his nearly seven years of employment with defendant, the company's own conduct and personnel policies gave rise to an "oral contract" not to fire him without good cause. The trial court sustained a demurrer without leave to amend on two grounds: that the complaint did not state facts sufficient to give rise to such contract, and that enforcement of any such contract would be barred by the statute of frauds. The Court of Appeal affirmed, relying on the latter ground alone. We consider both grounds, discussing the statute of frauds issue first.

A. Statute of Frauds Defense

Civil Code section 1624, subdivision (a), invalidates "[a]n agreement that by its terms is not to be performed within a year from the making thereof" unless the contract "or some note or memorandum thereof, [is] in writing and subscribed by the party to be charged or by the party's agent." [8a] In White Lighting Co. v. Wolfson (1968) 68 Cal. 2d 336 [66 Cal. Rptr. 697, 438 P.2d 345], we held that this portion of the statute of frauds "applies only to those contracts which, by their terms, cannot possibly be performed within one year." (Id., at p. 343.) In that case the employee alleged the breach of an express oral agreement whereby the defendant promised to employ him on a "permanent" basis and pay him a fixed commission on an [47 Cal. 3d 672] "annual basis." We concluded that the trial court erroneously sustained the defendant's demurrer because, although the agreement contemplated employment on a "permanent" basis, the statute does not apply to an employment contract of indefinite duration "unless its terms foreclose the employee's completion of the performance of the contract within one year ...." (Id., at p. 341.)

Relying exclusively on its own decision in Newfield v. Insurance Co. of the West (1984) 156 Cal. App. 3d 440 [203 Cal. Rptr. 9], the Court of Appeal here nevertheless held that plaintiff's alleged employment contract, if modified to include a promise to discharge him for cause only, is barred by the statute of frauds. Neither Newfield nor the opinion below distinguishes, or even cites, the rule in White Lighting, supra, 68 Cal. 2d 336. The rationale of both opinions is summed up in the following passage from Newfield: "[A]llegedly only [employee] had the right to terminate the contract. [9] Equality or justice between the parties would no longer exist in this alleged kind of oral contract. fn. [14] [¶] Appellant cannot have it both ways. Either his employment relationship was a contract in which both parties had equal rights to terminate at will (in which case it was not in violation of the statute of frauds), or it was a contract where the employer did not have the right to terminate at will, and there was a reasonable expectation of employment for more than one year (in which case the statute of frauds does apply, barring this action)." (156 Cal.App.3d at p. 446.) fn. 15

Newfield is irreconcilable with the rule in White Lighting. fn. 16 Even if the original oral agreement had expressly promised plaintiff "permanent" [47 Cal. 3d 673] employment terminable only on the condition of his subsequent poor performance or other good cause, such an agreement, if for no specified term, could possibly be completed within one year. (See White Lighting, supra, 68 Cal.2d at pp. 343-344.) Because the employee can quit or the employer can discharge for cause, even an agreement that strictly defines appropriate grounds for discharge can be completely performed within one year -- or within one day for that matter.

Our courts have consistently held that such contracts are not within the statute of frauds. (See, e.g., Plumlee v. Poag (1984) 150 Cal. App. 3d 541 [198 Cal. Rptr. 66]; Bondi v. Jewels by Edwar, Ltd. (1968) 267 Cal. App. 2d 672 [73 Cal. Rptr. 494]; Wescoatt v. Meeker (1944) 63 Cal. App. 2d 618 [147 P.2d 41]; Lloyd v. Kleefisch (1941) 48 Cal. App. 2d 408 [120 P.2d 97].) Decisions from other states uniformly hold that a good-cause termination clause does not render an employment agreement unenforceable under the statute of frauds. (E.g., Weiner v. McGraw-Hill, Inc. (1982) 57 N.Y.2d 458 [457 N.Y.S.2d 193, 443 N.E.2d 441, 33 A.L.R.4th 1101]; Toussaint v. Blue Cross & Blue Shield of Mich. (1980) 408 Mich. 579 [292 N.W.2d 880]; Hardison v. A.H. Belo Corp. (Tex.Civ.App. 1952) 247 S.W.2d 167.) [10] These authorities support the general rule that if a condition terminating a contract may occur within one year of its making, then the contract is performable within a year and does not fall within the scope of the statute of frauds. This is true even though performance of the contract may extend for longer than one year if the condition does not occur. (See generally,1 Witkin, Summary of Cal. Law (9th ed. 1987) Contracts, § 282, p. 274.)

Other courts have pointed out that within a year an employee such as plaintiff could have (1) been discharged for cause (see, e.g., Rowe v. Noren Pattern and Foundry Co. (1979) 91 Mich.App. 254 [283 N.W.2d 713]); (2) retired, died or voluntarily left employment (see, e.g., Martin v. Federal Life Ins. Co. (1982) 109 Ill.App.3d 596 [440 N.E.2d 998]); or (3) been terminated if declining profitability compelled a general layoff or cessation of business altogether (see, e.g., Stauter v. Walnut Grove Products (Iowa 1971) 188 N.W.2d 305). "Interpreting the allegations of the complaint liberally, as we must, we cannot say as a matter of law that the contract ... could not be [47 Cal. 3d 674] performed within a year." (Plumlee v. Poag, supra, 150 Cal. App. 3d 541, 549.)

Defendant attacks these precedents as performing "legalistic gymnastics," and calls instead for enforcement of Civil Code section 1624 according to the fair import of its language. That proclamation ignores a considerable piece of history. More than 60 years ago a British court declared that "[i]t is now two centuries too late to ascertain the meaning of [the statute of frauds] by applying one's own mind independently to the interpretation of its language. Our task is a much more humble one; it is to see how that section has been expounded in decisions and how the decisions apply to the present case." (Hanau v. Ehrlich (1911) 2 K.B. 1056, 1069.) fn. 17

The decision in White Lighting, supra, 68 Cal. 2d 336, follows a long line of precedent. In 1897, the Supreme Judicial Court of Massachusetts rejected an employer's contention that the statute of frauds invalidated an oral agreement for "permanent employment" so long as the plaintiff, an enameler, performed his work satisfactorily. The majority, including Chief Justice Field and Justice Holmes, rejected the employer's defense. "It has been repeatedly held that, if an agreement whose performance would otherwise extend beyond a year may be completely performed within a year on the happening of some contingency, it is not within the statute of frauds. [Citations.] In this case, we say nothing of other contingencies. The contract would have been completely performed if the defendant had ceased to carry on business within a year." (Carnig v. Carr (1897) 167 Mass. 544 [46 N.E. 117, 118].)

[8b] The Legislature, which in 1872 enacted Civil Code section 1624, was aware of precedent limiting the reach of the one-year provision of the statute of frauds. The Code Commissioners' notes specifically advised that "in a similar statute in New York these words have been construed as applying only to contracts which cannot possibly be executed within a year, under any contingency. ... To bring a contract within the statute relating to parol agreements, not to be performed within one year, it must appear to be necessarily incapable of performance within that time." (Italics in original.) fn. 18 [47 Cal. 3d 675] [11] When the Legislature enacts language that has received definitive judicial construction, we presume that the Legislature was aware of the relevant judicial decisions and intended to adopt that construction. (See Buchwald v. Katz (1972) 8 Cal. 3d 493, 502 [105 Cal. Rptr. 368, 503 P.2d 1376].) This presumption gains further strength when, as in this case, it is clear that the Legislature was explicitly informed of the prior construction.

[12] In sum, the contract between plaintiff and defendant could have been performed within one year of its making; plaintiff could have terminated his employment within that period, or defendant could have discharged plaintiff for cause. Thus, the contract does not fall within the statute of frauds and the fact that it was an implied or oral agreement is not fatal to its enforcement. fn. 19

B. Sufficiency of the Allegations of Oral or Implied Contract

[13] Although plaintiff describes his cause of action as one for breach of an oral contract, he does not allege explicit words by which the parties agreed that he would not be terminated without good cause. Instead he alleges that a course of conduct, including various oral representations, created a reasonable expectation to that effect. Thus, his cause of action is more properly described as one for breach of an implied-in-fact contract. fn. 20

As noted, the Court of Appeal did not reach the question of the sufficiency of plaintiff's allegations to state a cause of action for breach of an implied-in-fact contract term not to discharge except for good cause, because it disposed of the issue by erroneously applying the statute of frauds. [47 Cal. 3d 676] Nonetheless, the court extensively criticized the reasoning of Pugh, supra, 116 Cal. App. 3d 311, stating that it "destroys the centuries-old solid and settled principle of vast and demonstrated value to employer and employee, to the world of commerce and to the public of a contract which either can terminate at will." [14a] Before this court, defendant urges that we disapprove precedent permitting a cause of action for wrongful discharge founded on an implied-in-fact contract and require instead an express contract provision requiring good cause for termination, supported by independent consideration. Alternatively, defendant requests that we distinguish Pugh and its progeny from the present case. We conclude, however, that Pugh correctly applied basic contract principles in the employment context, and that these principles are applicable to plaintiff's agreement with defendant.

The plaintiff in Pugh had been employed by the defendant for 32 years, during which time he worked his way up the corporate ladder from dishwasher to vice president. (116 Cal.App.3d at p. 315.) When hired, he had been assured that "if you are loyal ... and do a good job, your future is secure." (Id., at p. 317.) During his long employment, the plaintiff received numerous commendations and promotions, and no significant criticism of his work. Throughout this period the company maintained a practice of not terminating administrative personnel without good cause. On this evidence, the Court of Appeal concluded that the jury could determine the existence of an implied promise that the employer would not arbitrarily terminate the plaintiff's employment. (Id., at p. 329.)

Pugh has been followed in numerous decisions in this state. (See, e.g., Steward v. Mercy Hospital, supra, 188 Cal. App. 3d 1290, 1296; Robinson v. Hewlett-Packard Corp. (1986) 183 Cal. App. 3d 1108, 1123 [228 Cal. Rptr. 591]; Hentzel v. Singer Co., supra, 138 Cal. App. 3d 290, 304 [same]; Walker v. Northern San Diego County Hospital Dist. (1982) 135 Cal. App. 3d 896, 904-905 [185 Cal. Rptr. 617]; cf. Shapiro v. Wells Fargo Realty Advisors, supra, 152 Cal. App. 3d 467, 479-482 [confirming but distinguishing Pugh, supra, 116 Cal. App. 3d 311].) A review of other jurisdictions also reveals a strong trend in favor of recognizing implied contract terms that modify the power of an employer to discharge an employee at will. (See, e.g., Hoffman-La Roche, Inc. v. Campbell (Ala. 1987) 512 So. 2d 725, 729-733 [implied promise to abide by employee manual guidelines contractually enforceable]; Thompson v. St. Regis Paper Co., supra, 685 P.2d 1081, 1087-1088 [if employer "creates an atmosphere of job security and fair treatment with promises of specific treatment in specific situations and an employee is induced thereby to remain on the job" those promises will be enforced (italics in original)]; Pine River State Bank v. Mettille (Minn. 1983) 333 N.W.2d 622, 626-629 [same]; Toussaint v. Blue Cross & Blue Shield of Mich., supra, 292 [47 Cal. 3d 677] N.W.2d 880, 892-893 ["contractual obligations can be implicit in employer policies and practices"]; see also Duldulao v. Saint Mary of Nazareth Hosp. (1987) 115 Ill. 2d 482 [505 N.E.2d 314, 317] [listing 25 jurisdictions recognizing contractually binding force of employee handbooks]; but see Sabetay v. Sterling Drug, Inc. (1987) 69 N.Y.2d 329 [514 N.Y.S.2d 209, 506 N.E.2d 919, 923] [requiring express agreement to limit employer's "unfettered right to terminate at will"].) Thus the rule defendant asks us to disapprove is one that has achieved widespread acceptance in recent years.

[15] We begin by acknowledging the fundamental principle of freedom of contract: employer and employee are free to agree to a contract terminable at will or subject to limitations. Their agreement will be enforced so long as it does not violate legal strictures external to the contract, such as laws affecting union membership and activity, prohibitions on indentured servitude, or the many other legal restrictions already described which place certain restraints on the employment arrangement. As we have discussed, Labor Code section 2922 establishes a presumption of at-will employment if the parties have made no express oral or written agreement specifying the length of employment or the grounds for termination. This presumption may, however, be overcome by evidence that despite the absence of a specified term, the parties agreed that the employer's power to terminate would be limited in some way, e.g., by a requirement that termination be based only on "good cause." (Drzewiecki v. H & R Block, Inc., supra, 24 Cal. App. 3d 695, 704; Millsap v. National Funding Corp. (1943) 57 Cal. App. 2d 772, 775-776 [135 P.2d 407].)

The absence of an express written or oral contract term concerning termination of employment does not necessarily indicate that the employment is actually intended by the parties to be "at will," because the presumption of at-will employment may be overcome by evidence of contrary intent. [16] Generally, courts seek to enforce the actual understanding of the parties to a contract, and in so doing may inquire into the parties' conduct to determine if it demonstrates an implied contract. "[I]t must be determined, as a question of fact, whether the parties acted in such a manner as to provide the necessary foundation for [an implied contract], and evidence may be introduced to rebut the inferences and show that there is another explanation for the conduct." (Silva v. Providence Hospital of Oakland (1939) 14 Cal. 2d 762, 774 [97 P.2d 798]; see Marvin v. Marvin (1976) 18 Cal. 3d 660, 684 [134 Cal. Rptr. 815, 557 P.2d 106]; Hillsman v. Sutter Community Hospitals (1984) 153 Cal. App. 3d 743, 753-756 [200 Cal. Rptr. 605]; Newberger v. Rifkind (1972) 28 Cal. App. 3d 1070, 1074 [104 Cal. Rptr. 663, 57 A.L.R.3d 1232]; 1 Witkin, Summary of Cal. Law, Contracts, op cit. supra, § 119, p. 145.) Such implied-in-fact contract terms ordinarily stand [47 Cal. 3d 678] on equal footing with express terms. (Rest.2d Contracts, supra, §§ 4, 19.) fn. 21 [14b] At issue here is whether the foregoing principles apply to contract terms establishing employment security, so that the presumption of Labor Code section 2922 may be overcome by evidence of contrary implied terms, or whether such agreements are subject to special substantive or evidentiary limitations.

Defendant contends that courts should not enforce employment security agreements in the absence of evidence of independent consideration and an express manifestation of mutual assent. Although, as explained below, there may be some historical basis for imposing such limitations, any such basis has been eroded by the development of modern contract law and, accordingly, we conclude that defendant's suggested limitations are inappropriate in the modern employment context. We discern no basis for departing from otherwise applicable general contract principles.

The doctrine that special limitations should be placed on the enforceability of employment security agreements arose during the late 19th century in the context of interpretation of contracts which promised "permanent" employment to the employee. In Lord v. Goldberg (1889) 81 Cal. 596, 601-602 [22 P. 1126], we held that language promising "permanent" employment created employment of no particular duration terminable only "for some good reason." fn. 22 Later cases developed the rule that "a contract for permanent employment, for life employment, for so long as the employee chooses, or for other terms indicating permanent employment, is interpreted as a contract for an indefinite period terminable at the will of either party ...." (See, e.g., Ruinello v. Murray (1951) 36 Cal. 2d 687, 689 [227 P.2d 251]; Speegle v. Board of Fire Underwriters (1946) 29 Cal. 2d 34, 39 [172 P.2d 867]; Shuler v. Corl (1918) 39 Cal. App. 195, 198 [178 P. 535].) This rule was not a substantive limitation on the parties' right to contract, but rather a rule of construction that could be overcome by a showing that the employment security term of the contract was supported by consideration other than the services to be rendered or by "other terms indicating a contrary intention." (Ruinello v. Murray, supra, 36 Cal.2d at p. 690; Drzewiecki v. H & R Block, Inc., supra, 24 Cal.App.3d at p. 704.) [47 Cal. 3d 679]

As the Pugh court explained, a rule imposing a requirement of separate consideration as a substantive limitation on an enforceable employee security agreement would be "contrary to the general contract principle that courts should not inquire into the adequacy of consideration. (See Calamari & Perillo, Contracts (2d ed. 1977) § 4-3, p. 136.) [17] 'A single and undivided consideration may be bargained for and given as the agreed equivalent of one promise or of two promises or of many promises.' (1 Corbin on Contracts (1963) § 125, pp. 535-536.) [14c] Thus there is no analytical reason why an employee's promise to render services, or his actual rendition of services over time, may not support an employer's promise both to pay a particular wage (for example) and to refrain from arbitrary dismissal. (See 1 Corbin on Contracts, op. cit. supra, § 125, p. 536, fn. 68; 1A Corbin on Contracts, op. cit. supra, § 152, pp. 13-17.)" (Pugh, supra, 116 Cal.App.3d at pp. 325-326, fn. omitted; see Drzewiecki v. H & R Block, Inc., supra, 24 Cal. App. 3d 695, 703-704; see also Pine River State Bank v. Mettille, supra, 333 N.W.2d at p. 629; Mauk, supra, 21 Idaho L. Rev. 201, 212-213; Note, Protecting At Will Employees, supra, 93 Harv. L. Rev. at pp. 1819-1820.)

The limitations on employment security terms on which defendant relies were developed during a period when courts were generally reluctant to look beyond explicit promises of the parties to a contract. "The court-imposed presumption that the employment contract is terminable at will relies upon the formalistic approach to contract interpretation predominant in late nineteenth century legal thought: manifestations of assent must be evidenced by definite, express terms if promises are to be enforceable." (Note, Protecting At Will Employees, supra, 93 Harv. L. Rev. at p. 1825, fns. omitted.) In the intervening decades, however, courts increasingly demonstrated their willingness to examine the entire relationship of the parties to commercial contracts to ascertain their actual intent, and this trend has been reflected in the body of law guiding contract interpretation. (See Goetz & Scott, The Limits of Expanded Choice: An Analysis of the Interactions Between Express and Implied Contract Terms (1985) 73 Cal.L.Rev. 261, 273-276 ["The (Uniform Commercial) Code, now joined by the Second Restatement of Contracts, effectively reverses the common law presumption that the parties' writing and the official law of contract are the definitive elements of the agreement. Evidence derived from experience and practice can now trigger the incorporation of additional, implied terms"].)

Similarly, 20 years ago, Professor Blumrosen observed that during the decades preceding his analysis, courts had demonstrated an increasing willingness to "consider the entire relationship of the parties, and to find that facts and circumstances establish a contract which cannot be terminated by the employer without cause." (Blumrosen, Settlement of Disputes Concerning the Exercise of Employer Disciplinary Power: United States Report, [47 Cal. 3d 680] supra, 18 Rutgers L.Rev. at p. 432, fn. omitted.) "This approach has been recognized as consistent with customary interpretation techniques of commercial contracts permitting 'gap filling' by implication of reasonable terms." (Miller & Estes, Recent Judicial Limitations on the Right to Discharge: A California Trilogy (1982) 16 U.C. Davis L.Rev. 65, 101, fn. omitted; see also McCarthy, Punitive Damages in Wrongful Discharge Cases (Lawpress 1985) §§ 3.55-3.56, pp. 206-207.)

[18] In the employment context, factors apart from consideration and express terms may be used to ascertain the existence and content of an employment agreement, including "the personnel policies or practices of the employer, the employee's longevity of service, actions or communications by the employer reflecting assurances of continued employment, and the practices of the industry in which the employee is engaged." (Pugh, supra, 116 Cal.App.3d at p. 327; see Note, Implied Contract Rights to Job Security (1974) 26 Stan.L.Rev. 335, 350-366 [reviewing factors courts have used in implied contract analyses].) Pursuant to Labor Code section 2922, if the parties reach no express or implied agreement to the contrary, the relationship is terminable at any time without cause. But when the parties have enforceable expectations concerning either the term of employment or the grounds or manner of termination, Labor Code section 2922 does not diminish the force of such contractual or legal obligations. fn. 23 The presumption that an employment relationship of indefinite duration is intended to be terminable at will is therefore "subject, like any presumption, to contrary evidence. This may take the form of an agreement, express or implied, that ... the employment relationship will continue indefinitely, pending the occurrence of some event such as the employer's dissatisfaction with the employee's services or the existence of some 'cause' for termination." (Pugh, supra, 116 Cal.App.3d at pp. 324-325, fn. omitted.)

[14d] Finally, we do not agree with the Court of Appeal that employment security agreements are so inherently harmful or unfair to employers, who do not receive equivalent guaranties of continued service, as to merit [47 Cal. 3d 681] treatment different from that accorded other contracts. On the contrary, employers may benefit from the increased loyalty and productivity that such agreements may inspire. (See Mauk, supra, 21 Idaho L. Rev. 201, 217.) Permitting proof of and reliance on implied-in-fact contract terms does not nullify the at-will rule, it merely treats such contracts in a manner in keeping with general contract law. (See Note, Defining Public Policy Torts in At-Will Dismissals (1981) 34 Stan.L.Rev. 153, 154-155 [hereafter Defining Torts] ["While the implied contract approach reflects a movement away from the harshness of the at-will rule, it by no means represents a rejection of the rule, since it merely allows employees to rebut more easily the presumption that their employment is terminable at will" (fn. omitted)].) We see no sound reason to exempt the employment relationship from the ordinary rules of contract interpretation which permit proof of implied terms.

Defendant's remaining argument is that even if a promise to discharge "for good cause only" could be implied in fact, the evidentiary factors outlined in Pugh, supra, 116 Cal.App.3d at page 329, and relied on by plaintiff, are inadequate as a matter of law. This contention fails on several grounds.

First, defendant overemphasizes the fact that plaintiff was employed for "only" six years and nine months. Length of employment is a relevant consideration but six years and nine months is sufficient time for conduct to occur on which a trier of fact could find the existence of an implied contract. (Cf. Gray v. Superior Court, supra, 181 Cal. App. 3d 813, 821; Khanna v. Microdata Corp. (1985) 170 Cal. App. 3d 250, 262 [215 Cal. Rptr. 860].) [19] As to establishing the requisite promise, "oblique language will not, standing alone, be sufficient to establish agreement"; instead, the totality of the circumstances determines the nature of the contract. Agreement may be "'shown by the acts and conduct of the parties, interpreted in the light of the subject matter and of the surrounding circumstances.'" (Pugh, supra, 116 Cal.App.3d at p. 329.) [14e] Plaintiff here alleged repeated oral assurances of job security and consistent promotions, salary increases and bonuses during the term of his employment contributing to his reasonable expectation that he would not be discharged except for good cause.

[20] Second, an allegation of breach of written "Termination Guidelines" implying self-imposed limitations on the employer's power to discharge at will may be sufficient to state a cause of action for breach of an employment contract. Pugh, supra, 116 Cal. App. 3d 311, is not alone in holding that the trier of fact can infer an agreement to limit the grounds for termination based on the employee's reasonable reliance on the company's [47 Cal. 3d 682] personnel manual or policies. (See, e.g., Robinson v. Hewlett-Packard Corp., supra, 183 Cal.App.3d at p. 1123 [promise not to terminate without good cause demonstrated by personnel guidelines and individual performance warnings, evaluations and instructions]; Rulon-Miller v. International Business Machines Corp. (1984) 162 Cal. App. 3d 241, 251 [208 Cal. Rptr. 524] [factual issue whether termination was for reasons in stated employer policies]; Walker v. Northern San Diego County Hospital Dist., supra, 135 Cal.App.3d at pp. 904-905 [handbook creating right to discharge only for cause and to pretermination hearing]; Toussaint v. Blue Cross & Blue Shield of Mich., supra, 292 N.W.2d at p. 892 [personnel manual provisions can give rise to contractual rights without showing of express mutual agreement]; Morris v. Lutheran Medical Center (1983) 215 Neb. 677 [340 N.W.2d 388, 390-391] [employer bound by published "Policy and Procedures"]; cf. Hepp v. Lockheed-California Co. (1978) 86 Cal. App. 3d 714, 719 [150 Cal. Rptr. 408] [unwritten but "well established" policy regulating rehiring of employees laid off for lack of work is enforceable].)

[14f] Finally, unlike the employee in Pugh, supra, 116 Cal. App. 3d 311, plaintiff alleges that he supplied the company valuable and separate consideration by signing an agreement whereby he promised not to compete or conceal any computer-related information from defendant for one year after termination. The noncompetition agreement and its attendant "Disclosure and Assignment of Proprietary Information, Inventions, etc." may be probative evidence that "it is more probable that the parties intended a continuing relationship, with limitations upon the employer's dismissal authority [because the] employee has provided some benefit to the employer, or suffers some detriment, beyond the usual rendition of service." (Pugh, supra, 116 Cal.App.3d at p. 326.)

In sum, plaintiff has pleaded facts which, if proved, may be sufficient for a jury to find an implied-in-fact contract limiting defendant's right to discharge him arbitrarily -- facts sufficient to overcome the presumption of Labor Code section 2922. On demurrer, we must assume these facts to be true. In other words, plaintiff has pleaded an implied-in-fact contract and its breach, and is entitled to his opportunity to prove those allegations. fn. 24

III. Breach of the Implied Covenant of Good Faith and Fair Dealing

We turn now to plaintiff's cause of action for tortious breach of the implied covenant of good faith and fair dealing. [21a] Relying on Cleary, [47 Cal. 3d 683] supra, 111 Cal. App. 3d 443, and subsequent Court of Appeal cases, plaintiff asserts we should recognize tort remedies for such a breach in the context of employment termination.

[22] The distinction between tort and contract is well grounded in common law, and divergent objectives underlie the remedies created in the two areas. Whereas contract actions are created to enforce the intentions of the parties to the agreement, tort law is primarily designed to vindicate "social policy." (Prosser, Law of Torts (4th ed. 1971) p. 613.) The covenant of good faith and fair dealing was developed in the contract arena and is aimed at making effective the agreement's promises. Plaintiff asks that we find that the breach of the implied covenant in employment contracts also gives rise to an action seeking an award of tort damages.

In this instance, where an extension of tort remedies is sought for a duty whose breach previously has been compensable by contractual remedies, it is helpful to consider certain principles relevant to contract law. First, predictability about the cost of contractual relationships plays an important role in our commercial system. (Putz & Klippen, Commercial Bad Faith: Attorney Fees -- Not Tort Liability -- Is the Remedy for "Stonewalling" (1987) 21 U.S.F. L.Rev. 419, 432 [hereafter Putz & Klippen]). Moreover, "Courts traditionally have awarded damages for breach of contract to compensate the aggrieved party rather than to punish the breaching party." (Note, "Contort": Tortious Breach of the Implied Covenant of Good Faith and Fair Dealing in Noninsur- ance, Commercial Contracts -- Its Existence and Desirability (1985) 60 Notre Dame L. Rev. 510, 526, & fn. 94, citing Rest.2d Contracts, § 355, com. a ["The purpose[] of awarding contract damages is to compensate the injured party"].) fn. 25 With these concepts in mind, we turn to analyze the role of the implied covenant of good faith and fair dealing and the propriety of the extension of remedies urged by plaintiff.

[23] "Every contract imposes upon each party a duty of good faith and fair dealing in its performance and its enforcement." (Rest.2d Contracts, § 205.) This duty has been recognized in the majority of American jurisdictions, the Restatement, and the Uniform Commercial Code. (Burton, Breach of Contract and the Common Law Duty to Perform in Good Faith [47 Cal. 3d 684] (1980) 94 Harv. L. Rev. 369.) Because the covenant is a contract term, however, compensation for its breach has almost always been limited to contract rather than tort remedies. As to the scope of the covenant, "'[t]he precise nature and extent of the duty imposed by such an implied promise will depend on the contractual purposes." (Egan v. Mutual of Omaha Ins. Co. (1979) 24 Cal. 3d 809, 818 [169 Cal. Rptr. 691, 620 P.2d 141].) Initially, the concept of a duty of good faith developed in contract law as "a kind of 'safety valve' to which judges may turn to fill gaps and qualify or limit rights and duties otherwise arising under rules of law and specific contract language." (Summers, The General Duty of Good Faith -- Its Recognition and Conceptualization (1982) 67 Cornell L. Rev. 810, 812, fn. omitted; see also Burton, supra, 94 Harv. L. Rev. 369, 371 ["the courts employ the good faith doctrine to effectuate the intentions of parties, or to protect their reasonable expectations" (fn. omitted)].) As a contract concept, breach of the duty led to imposition of contract damages determined by the nature of the breach and standard contract principles.

[24] An exception to this general rule has developed in the context of insurance contracts where, for a variety of policy reasons, courts have held that breach of the implied covenant will provide the basis for an action in tort. California has a well-developed judicial history addressing this exception. In Comunale v. Traders & General Ins. Co. (1958) 50 Cal. 2d 654, 658 [328 P.2d 198, 68 A.L.R.2d 883], we stated, "There is an implied covenant of good faith and fair dealing in every contract that neither party will do anything which will injure the right of the other to receive the benefits of the agreement." (See also Egan v. Mutual of Omaha Ins. Co., supra, 24 Cal. 3d 809, 818.) Thereafter, in Crisci v. Security Ins. Co. (1967) 66 Cal. 2d 425 [58 Cal. Rptr. 13, 426 P.2d 173], for the first time we permitted an insured to recover in tort for emotional damages caused by the insurer's breach of the implied covenant. We explained in Gruenberg v. Aetna Ins. Co. (1973) 9 Cal. 3d 566 [108 Cal. Rptr. 480, 510 P.2d 1032], that "[t]he duty [to comport with the implied covenant of good faith and fair dealing] is immanent in the contract whether the company is attending [on the insured's behalf] to the claims of third persons against the insured or the claims of the insured itself. Accordingly, when the insurer unreasonably and in bad faith withholds payment of the claim of its insured, it is subject to liability in tort." (Id., at p. 575.)

In Egan v. Mutual of Omaha Ins. Co., supra, 24 Cal. 3d 809, we described some of the bases for permitting tort recovery for breach of the implied covenant in the insurance context. "The insured in a contract like the one before us does not seek to obtain a commercial advantage by purchasing the policy -- rather, he seeks protection against calamity." (Id., at p. 819.) Thus, "As one commentary has noted, 'The insurers' obligations are ... rooted [47 Cal. 3d 685] in their status as purveyors of a vital service labeled quasi-public in nature. Suppliers of services affected with a public interest must take the public's interest seriously, where necessary placing it before their interest in maximizing gains and limiting disbursements. ... [A]s a supplier of a public service rather than a manufactured product, the obligations of insurers go beyond meeting reasonable expectations of coverage. The obligations of good faith and fair dealing encompass qualities of decency and humanity inherent in the responsibilities of a fiduciary.' ... (Goodman & Seaton, Foreword: Ripe for Decision, Internal Workings and Current Concerns of the California Supreme Court (1974) 62 Cal.L.Rev. 309, 346-347.)" (24 Cal.3d at p. 820.)

In addition, the Egan court emphasized that "the relationship of insurer and insured is inherently unbalanced: the adhesive nature of insurance contracts places the insurer in a superior bargaining position." (24 Cal.3d at p. 820.) This emphasis on the "special relationship" of insurer and insured has been echoed in arguments and analysis in subsequent scholarly commentary and cases which urge the availability of tort remedies in the employment context.

The first California appellate case to permit tort recovery in the employment context was Cleary, supra, 111 Cal. App. 3d 443. To support its holding that tort as well as contract damages were appropriate to compensate for a breach of the implied covenant, the Cleary court relied on insurance cases (Egan, supra, 24 Cal. 3d 809, and Comunale, supra, 50 Cal.2d 654) without engaging in comparative analysis of insurance and employment relationships and without inquiring into whether the insurance cases' departure from established principles of contract law should generally be subject to expansion.

Similarly, Cleary's discussion of two previous California employment cases was insufficient. It found a "hint" in Coats v. General Motors Corp. (1934) 3 Cal. App. 2d 340, 348 [39 P.2d 838], to support the proposition that "on occasion, it may be incumbent upon an employer to demonstrate good faith in terminating an employee" (111 Cal.App.3d at p. 453), but failed to acknowledge that in Coats, the employee sought recovery of only contract damages (Coats, supra, 3 Cal.App.2d at pp. 344-345; see also Zimmer v. Wells Management Corp. (S.D.N.Y. 1972) 348 F. Supp. 540 [employment case, also cited in Cleary, which did not involve tort damages]). Next, the Cleary court placed undue reliance on dictum in this court's Tameny decision, supra, 27 Cal.3d at page 179, footnote 12, which suggested that tort remedies might be available when an employer breaches the implied covenant of good faith and fair dealing. (111 Cal.App.3d at pp. 454-455.) The qualified Tameny dictum was based exclusively on precedent in insurance [47 Cal. 3d 686] cases from this state, and two out-of-state employment cases. The out-of-state cases included Monge v. Beebe Rubber Company (1974) 114 N.H. 130 [316 A.2d 549, 62 A.L.R.3d 264], in which the court permitted an action for wrongful discharge but limited the plaintiff's recovery to contract damages, specifically excluding recovery for mental distress. (Id., at pp. 551-552.) Moreover, the New Hampshire Supreme Court thereafter confined Monge to cases in which the employer's actions contravene public policy. (Howard v. Dorr Woolen Co. (1980) 120 N.H. 295 [414 A.2d 1273, 1274].) In the second case, Fortune v. National Cash Register Co. (1977) 373 Mass. 96 [364 N.E.2d 1251, 1256], the court created a right of action based on breach of the implied covenant, but limited recovery to benefits the employee had already earned under the contract. Subsequent Massachusetts cases have pursued the same limited course. (See, e.g., Maddaloni v. Western Mass. Bus Lines, Inc. (1982) 386 Mass. 877 [438 N.E.2d 351, 356] [if covenant of good faith and fair dealing breached, employee limited to benefits contemplated or included in contract].)

In fact, although Justice Broussard asserts that the weight of authority is in favor of granting a tort remedy (opn. by Broussard, J., at p. 714, fn. 15), the clear majority of jurisdictions have either expressly rejected the notion of tort damages for breach of the implied covenant in employment cases or impliedly done so by rejecting any application of the covenant in such a context. fn. 26 [47 Cal. 3d 687]

Both the Tameny dictum and Cleary, supra, 111 Cal. App. 3d 443, failed to recognize that imposing tort liability for breach of the implied covenant was unprecedented in the employment area. The Tameny court expressly found it unnecessary to venture into the area of the implied covenant more completely (27 Cal.3d at p. 179, fn. 12) and only briefly touched on the subject. The Cleary court erred in its uncritical reliance on insurance law and its casual extension of Tameny to find tort damages recoverable in the case before it.

Dictum in Seaman's Direct Buying Service, Inc. v. Standard Oil Co. (1984) 36 Cal. 3d 752 [206 Cal. Rptr. 354, 686 P.2d 1158] also is not helpful. There, the court focused on a standard commercial contract. We stated, "[w]hile the proposition that the law implies a covenant of good faith and fair dealing in all contracts is well established, the proposition advanced by Seaman's -- that breach of the covenant always gives rise to an action in tort -- is not so clear." (Id., at p. 768.) We also observed that the propriety of a tort action for breach of the implied covenant in the insurance context was based on the "special relationship" of insurer and insured, and continued, "No doubt there are other relationships with similar characteristics and deserving of similar legal treatment." (Id., at pp. 768-769.) In a footnote to the last statement, we referred to Tameny, observing that there "this court intimated that breach of the covenant of good faith and fair dealing in the employment relationship might give rise to tort remedies. That relationship has some of the same characteristics as the relationship between insurer and [47 Cal. 3d 688] insured." (Id., at p. 769, fn. 6.) This allusion to the potential for extending tort remedies for breach of the implied covenant was tentative at best. fn. 27

Most of the other Court of Appeal cases following Cleary suffer from similar failures comprehensively to consider the implications of their holdings. These opinions either merely refused to find a breach of the implied covenant on the facts of the case, or relied uncritically on Cleary or the dicta in Tameny and Seaman's. (See, e.g., Gray v. Superior Court, supra, 181 Cal. App. 3d 813, 820-821; Rulon-Miller v. International Business Machines Corp., supra, 162 Cal.App.3d at pp. 252-253; Shapiro v. Wells Fargo Realty Advisors, supra, 152 Cal.App.3d at pp. 478-479; Crosier v. United Parcel Service, Inc. (1983) 150 Cal. App. 3d 1132, 1137 [198 Cal. Rptr. 361]; cf. Wayte v. Rollins International, Inc. (1985) 169 Cal. App. 3d 1, 20 [215 Cal. Rptr. 59].) For example, the Court of Appeal decisions in Koehrer v. Superior Court, supra, 181 Cal. App. 3d 1155, 1167-1171 and Khanna v. Microdata Corp., supra, 170 Cal. App. 3d 250, 260-262, although engaging in some analysis, proceed from certain unexplored assumptions about the role of the covenant of good faith and fair dealing and the circumstances in which tort damages may be found permissible. The Ninth Circuit recently followed the path taken in the Koehrer and Khanna decisions, stating that their approach "is more compelling because it does not treat the implied covenant of good faith and fair dealing in employment contracts as a distinct area of the law." (Huber v. Standard Ins. Co. (9th Cir. 1988) 841 F.2d 980, 984.) The Huber court, however, did not critically examine the underpinnings of the Court of Appeal decisions.

In Koehrer, supra, 181 Cal. App. 3d 1155, the court acknowledged that we found it unnecessary to base our decision in Seaman's on the implied covenant of good faith and fair dealing, but nonetheless concluded that we essentially had done so. Despite the fact that we carefully limited our holding in Seaman's to instances in which a party "seeks to shield itself from liability by denying, in bad faith and without probable cause, that the contract exists" (36 Cal.3d at p. 769), the Koehrer court expansively concluded that, "If ... the existence of good cause for discharge is asserted by the employer without probable cause and in bad faith, that is, without a good faith belief that good cause for discharge in fact exists, the employer has tortiously attempted to deprive the employee of the benefits of the agreement, and an action for breach of the implied covenant of good faith and fair dealing will lie." (181 Cal.App.3d at p. 1171.) Koehrer thus extended the expressly circumscribed cause of action established in Seaman's [47 Cal. 3d 689] based on denial of the existence of the contract, to find a tort cause of action when the dispute related to a contract term, namely the necessity for good cause as a basis for termination. By this broad stroke, made without analyzing the appropriateness of imposing tort remedies in the employment context, the Koehrer court broached the possibility of obtaining tort damages for the breach of any term of a contract whether for employment or otherwise.

Similarly, in Khanna, supra, 170 Cal. App. 3d 250, the court used Tameny and Cleary as its starting point and focused primarily on the appropriate factual grounds on which an action for breach of the covenant could be brought without analyzing the propriety of the remedy itself. The court concluded that the correct inquiry was whether there was sufficient evidence to support a finding that the employer "engaged in bad faith action, extraneous to the contract, with the motive intentionally to frustrate respondent's enjoyment of his contract rights." (Id., at p. 263.) It did not, however, focus on the fact that traditionally such a finding justified only contract damages. Finally, Huber, supra, 841 F.2d at pages 983-985, engaged in no additional exploration of the justification for tort remedies, simply selecting Koehrer and Khanna as presenting the better of two approaches to the necessary elements of the tort described in California cases.

[25] In our view, the underlying problem in the line of cases relied on by plaintiff lies in the decisions' uncritical incorporation of the insurance model into the employment context, without careful consideration of the fundamental policies underlying the development of tort and contract law in general or of significant differences between the insurer/insured and employer/employee relationships. fn. 28 When a court enforces the implied covenant it is in essence acting to protect "the interest [47 Cal. 3d 690] in having promises performed" (Prosser, Law of Torts, op cit. supra, p. 613) -- the traditional realm of a contract action -- rather than to protect some general duty to society which the law places on an employer without regard to the substance of its contractual obligations to its employee. Thus, in Tameny, 27 Cal. 3d 167, 175-176, as we have explained, the court was careful to draw a distinction between "ex delicto" and "ex contractu" obligations. (See, ante, at pp. 667-668.) [26] An allegation of breach of the implied covenant of good faith and fair dealing is an allegation of breach of an "ex contractu" obligation, namely one arising out of the contract itself. The covenant of good faith is read into contracts in order to protect the express covenants or promises of the contract, not to protect some general public policy interest not directly tied to the contract's purposes. The insurance cases thus were a major departure from traditional principles of contract law. We must, therefore, consider with great care claims that extension of the exceptional approach taken in those cases is automatically appropriate if certain hallmarks and similarities can be adduced in another contract setting. With this emphasis on the historical purposes of the covenant of good faith and fair dealing in mind, we turn to consider the bases upon which extension of the insurance model to the employment sphere has been urged.

[21c] The "special relationship" test gleaned from the insurance context has been suggested as a model for determining the appropriateness of permitting tort remedies for breach of the implied covenant of the employment context. One commentary has observed, "[j]ust as the law of contracts fails to provide adequate principles for construing the terms of an insurance policy, the substantial body of law uniquely applicable to insurance contracts is practically irrelevant to commercially oriented contracts. ... These [unique] features characteristic of the insurance contract make it particularly susceptible to public policy considerations." (Louderback & Jurika, Standards for Limiting the Tort of Bad Faith Breach of Contract (1982) 16 U.S.F. L.Rev. 187, 200-201, fns. omitted.) These commentators assert that tort remedies for breach of the covenant should not be extended across the board in the commercial context, but that, nonetheless, public policy considerations suggest extending the tort remedy if certain salient factors are present. (Id., at pp. 216-218.) "The tort of bad faith should be applied to commercial contracts only if four of the features characteristic of insurance bad faith actions are present. The features are: (1) one of the parties to the contract enjoys a superior bargaining position to the extent that it is able to dictate the terms of the contract; (2) the purpose of the weaker party in entering into the contract is not primarily to profit but rather to secure an essential service or product, financial security or peace of mind; (3) the relationship of the parties is such that the weaker party places its trust and confidence in the larger entity; and (4) there is conduct on the [47 Cal. 3d 691] part of the defendant indicating an intent to frustrate the weaker party's enjoyment of the contract rights." (Id., at p. 227.) The discussion of these elements includes an assumption that a tort remedy should be recognized in employment relationships within the stated limitations. fn. 29

Others argue that the employment context is not sufficiently analogous to that of insurance to warrant recognition of the right to tort recovery. (See, e.g., Miller & Estes, supra, 16 U.C. Davis L.Rev. 65, 90-91; Note, Defining Torts, supra, 34 Stan.L.Rev. at pp. 164-167.) They contend that (1) inequality in bargaining power is not a universal characteristic of employment contracts, standardized forms are often not used, and there is often room for bargaining as to special conditions; (2) employers do not owe similar fiduciary duties to employees who are themselves agents of the employer and obligated to act in the employer's interests; and (3) unlike insurance companies, employers are not "quasi-public entities" and they "seldom have government-like functions, and do not serve primarily, if at all, to spread losses across society." (Note, Defining Torts, supra, 34 Stan.L.Rev. at pp. 165-167; Miller & Estes, supra, 16 U.C. Davis L.Rev. at p. 91.)

In contrast to those concentrating on the match between insurance and employment relationships, yet another article suggests, "The fundamental flaw in the 'special relationship' test is that it is illusory. It provides a label to hang on a result but not a principled basis for decision. ... The qualifying contracts cannot be identified until the issue has been litigated, which is too late." (Putz & Klippen, supra, 21 U.S.F. L.Rev. at pp. 478-479.) The authors assert that "'public interest, adhesion and fiduciary responsibility,' are not sufficiently precise to provide a basis for reliable prediction." (Id., at p. 479, fn. omitted.) Instead, they assert that, "While the 'special relationship' test purports to be only a modest extension of the tort of bad faith beyond insurance and employment, it opens the way for pleading a tort cause of action in nearly every contract case, leaving it ultimately to a jury to decide whether or not the parties had a 'special relationship.'" (Id., at p. 480, fn. omitted.) Extension of the test to employment cases would similarly leave the door open to such a claim in every termination case, and [47 Cal. 3d 692] readers are cautioned not to infer "that the authors support extension of tort liability beyond insurance through use of the 'special relationship' test." (Id., at p. 461, fn. 163.)

Similarly, another commentary argues that the special relationship model fails because (1) it does not explain why it "justifies tort liability" for otherwise legal conduct, or for conduct which may give rise to contract remedies (Comment, Reconstructing Breach, supra, 73 Cal.L.Rev. at p. 1299, fn. omitted), (2) use of the concept "is inadequate to define the scope and application of a tort duty of good faith and fair dealing" (id., at p. 1300), (3) use of the model "fails to distinguish between breach of the implied covenant of good faith and fair dealing and 'bad faith breach of contract'" (ibid., fn. omitted), and (4) the model does not provide justification for imposition of punitive damages and thus "might serve to unfairly chill legitimate conduct" (id., at p. 1301).

After review of the various commentators, and independent consideration of the similarities between the two areas, we are not convinced that a "special relationship" analogous to that between insurer and insured should be deemed to exist in the usual employment relationship which would warrant recognition of a tort action for breach of the implied covenant. Even if we were to assume that the special relationship model is an appropriate one to follow in determining whether to expand tort recovery, a breach in the employment context does not place the employee in the same economic dilemma that an insured faces when an insurer in bad faith refuses to pay a claim or to accept a settlement offer within policy limits. When an insurer takes such actions, the insured cannot turn to the marketplace to find another insurance company willing to pay for the loss already incurred. [27] , [21d] The wrongfully terminated employee, on the other hand, can (and must, in order to mitigate damages [see Parker v. Twentieth Century-Fox Film Corp. (1970) 3 Cal. 3d 176, 181-182 [89 Cal. Rptr. 737, 474 P.2d 689, 44 A.L.R.3d 615]]) make reasonable efforts to seek alternative employment. (See Mauk, supra, 21 Idaho L.Rev. 201, 208.) Moreover, the role of the employer differs from that of the "quasi-public" insurance company with whom individuals contract specifically in order to obtain protection from potential specified economic harm. The employer does not similarly "sell" protection to its employees; it is not providing a public service. Nor do we find convincing the idea that the employee is necessarily seeking a different kind of financial security than those entering a typical commercial contract. If a small dealer contracts for goods from a large supplier, and those goods are vital to the small dealer's business, a breach by the supplier may have financial significance for individuals employed by the dealer or to the dealer himself. Permitting only contract damages in [47 Cal. 3d 693] such a situation has ramifications no different from a similar limitation in the direct employer-employee relationship.

Finally, there is a fundamental difference between insurance and employment relationships. In the insurance relationship, the insurer's and insured's interest are financially at odds. If the insurer pays a claim, it diminishes its fiscal resources. The insured, of course, has paid for protection and expects to have its losses recompensed. When a claim is paid, money shifts from insurer to insured, or, if appropriate, to a third party claimant.

Putting aside already specifically barred improper motives for termination which may be based on both economic and noneconomic considerations, fn. 30 as a general rule it is to the employer's economic benefit to retain good employees. The interests of employer and employee are most frequently in alignment. If there is a job to be done, the employer must still pay someone to do it. This is not to say that there may never be a "bad motive" for discharge not otherwise covered by law. Nevertheless, in terms of abstract employment relationships as contrasted with abstract insurance relationships, there is less inherent relevant tension between the interests of employers and employees than exists between that of insurers and insureds. Thus the need to place disincentives on an employer's conduct in addition to those already imposed by law simply does not rise to the same level as that created by the conflicting interests at stake in the insurance context. Nor is this to say that the Legislature would have no basis for affording employees additional protections. It is, however, to say that the need to extend the special relationship model in the form of judicially created relief of the kind sought here is less compelling.

We therefore conclude that the employment relationship is not sufficiently similar to that of insurer and insured to warrant judicial extension of the proposed additional tort remedies in view of the countervailing concerns about economic policy and stability, the traditional separation of tort and contract law, and finally, the numerous protections against improper terminations already afforded employees.

Our inquiry, however, does not end here. The potential effects on an individual caused by termination of employment arguably justify additional remedies for certain improper discharges. The large body of employment [47 Cal. 3d 694] law restricting an employer's right to discharge based on discriminatory reasons or on the employee's exercise of legislatively conferred employee rights, indicates that the Legislature and Congress have recognized the importance of the employment relationship and the necessity for vindication of certain legislatively and constitutionally established public policies in the employment context. The Tameny cause of action likewise is responsive to similar public concerns. In the quest for expansion of remedies for discharged workers which we consider here, however, the policies sought to be vindicated have a different origin. The most frequently cited reason for the move to extend tort remedies in this context is the perception that traditional contract remedies are inadequate to compensate for certain breaches. (See, e.g., Putz & Klippen, supra, 21 U.S.F. L.Rev. at pp. 470-471; Traynor, Bad Faith Breach of a Commercial Contract: A Comment on the Seaman's Case (Cal. State Bar, Fall 1984) 8 Bus. L. News 1.) Others argue that the quest for additional remedies specifically for terminated workers also has its genesis in (1) comparisons drawn between the protections afforded nonunion employees and those covered by collective bargaining agreements, (2) changes in the economy which have led to displacement of middle-level management employees in "unprecedented numbers," and (3) the effect of antidiscrimination awareness and legislation that has "raised expectations and created challenges to employer decision making." (Gould, Stemming the Wrongful Discharge Tide: A Case for Arbitration (1988) 13 Emp.Rel.L.J. 404, 408-410 [hereafter Stemming the Tide].)

The issue is how far courts can or should go in responding to these concerns regarding the sufficiency of compensation by departing from long established principles of contract law. Significant policy judgments affecting social policies and commercial relationships are implicated in the resolution of this question in the employment termination context. Such a determination, which has the potential to alter profoundly the nature of employment, the cost of products and services, and the availability of jobs, arguably is better suited for legislative decisionmaking. (See Wagenseller v. Scottsdale Memorial Hosp., supra, 710 P.2d 1025, 1040; Gould, The Idea of the Job as Property in Contemporary America: The Legal and Collective Bargaining Framework, 1986 B.Y.U. L. Rev. 885, 898, 908 [hereafter The Idea of the Job]; cf. Sabetay v. Sterling Drug, Inc., supra, 506 N.E.2d at p. 923.) Moreover, as we discuss, the extension of the availability of tort remedies is but one among many solutions posited to remedy the problem of adequately compensating employees for certain forms of "wrongful" termination while balancing the interests of employers in their freedom to make economically based decisions about their work force. fn. 31 [47 Cal. 3d 695]

It cannot be disputed that legislation at both the state and national level has profoundly affected the scope of at-will terminations. As noted, regulation of employment ranging from workers' compensation laws to antidiscrimination enactments, fair labor standards, minimum compensation, regulation of hours, etc., all have significantly impinged on the laissez-faire underpinnings of the at-will rule. (See, ante, p. 665, fn. 4.) Moreover, unionization of a portion of the domestic workforce has substantial implications for the judicial development of employment termination law because the rights of such workers when terminated are often governed exclusively by the terms of applicable collective bargaining agreements. fn. 32 The slate we write on thus is far from clean.

Professor Gould asserts, "[t]he new common law of wrongful discharge has provided employer and employee with the worst of all possible worlds. ... [E]mployers are subject to volatile and unpredictable juries that frequently act without regard to legal instructions. Moreover, the employees who benefit are few and far between, first, because of the difficulties involved in staying the course of a lengthy and expensive judicial process, and second, because of limitations inherent in the legal doctrines adopted by [47 Cal. 3d 696] the courts." (Gould, Stemming the Tide, supra, 13 Emp.Rel.L.J. at p. 413.) Gould advocates exploring arbitration as an alternative, and his emphasis on the sporadic effectiveness of the tort cause of action to remedy perceived inadequacies in employee protection is important to our consideration of the effectiveness of the remedy sought here.

Professor Putz and coauthor Klippen also suggest "tort liability is not the answer to bad faith defense in commercial contract disputes. A more appropriate response is to make contract damages adequate by permitting a prevailing plaintiff to recover attorney fees where the breaching party is found to have denied liability unreasonably." (Putz & Klippen, supra, 21 U.S.F. L.Rev. at p. 499.) Yet another commentator advocates expansion of recoverable contract damages. (Traynor, supra, 8 Bus.L.News at pp. 12-14.) Others would permit tort damages but would limit their application. (See, e.g., Louderback & Jurika, supra, 16 U.S.F. L.Rev. at pp. 220-223; Comment, Reconstructing Breach, supra, 73 Cal.L.Rev. at pp. 1315-1330.) These various approaches on the one hand suggest a widespread perception that present compensation is inadequate, but on the other hand vividly demonstrate substantial disagreement about the propriety or even the potential form of tort remedies for breaches of contractual duties of covenants. The multiplicity of solutions advanced underscores the caution with which any attempts to extend such relief must be viewed.

As we have reiterated, the employment relationship is fundamentally contractual, and several factors combine to persuade us that in the absence of legislative direction to the contrary contractual remedies should remain the sole available relief for breaches of the implied covenant of good faith and fair dealing in the employment context. Initially, predictability of the consequences of actions related to employment contracts is important to commercial stability. fn. 33 In order to achieve such stability, it is also important that employers not be unduly deprived of discretion to dismiss an employee by the fear that doing so will give rise to potential tort recovery in every case. [47 Cal. 3d 697]

Moreover, it would be difficult if not impossible to formulate a rule that would assure that only "deserving" cases give rise to tort relief. Professor Summers, in his seminal article, described the term "good faith" as used in the duty of good faith imposed in contract law and the Uniform Commercial Code, fn. 34 as an "excluder" phrase which is "without general meaning (or meanings) of its own and serves to exclude a wide range of heterogenous forms of bad faith. In a particular context the phrase takes on specific meaning, but usually this is only by way of contrast with the specific form of bad faith actually or hypothetically ruled out." (Summers, "Good Faith" in General Contract Law and the Sales Provisions of the Uniform Commercial Code (1968) 54 Va. L. Rev. 195, 201, fn. omitted.) In a tort action based on an employee's discharge, it is highly likely that each case would involve a dispute as to material facts regarding the subjective intentions of the employer. fn. 35 As a result, these actions could rarely be disposed of at the demurrer or summary judgment stage.

The formulation advanced in Koehrer, supra, 181 Cal.App.3d at page 1171 (see ante, p. 688), affords no real restriction on the employee's ability to bring an action after termination. fn. 36 Nor did the Khanna court's approach [47 Cal. 3d 698] provide such necessary delineation. The court provided simply: "A breach of the implied covenant of good faith and fair dealing in employment contracts is established whenever the employer engages in 'bad faith action extraneous to the contract, combined with the obligor's intent to frustrate the [employee's] enjoyment of contract rights.' The facts in Cleary [supra, 111 Cal. App. 3d 443] establish only one manner among many by which an employer might violate this covenant." (170 Cal.App.3d at p. 262, see also Huber, supra, 841 F.2d at p. 985 [discussing necessary showing for a prima facie case: employee need only show "unjust termination" but need not show bad or hidden motivation on employer's part].) fn. 37

Review of the Koehrer, Khanna and Huber formulations reveals that ultimately they require nothing "unusual" about the breach: under the approaches of those courts, an ordinary contract breach might give rise to a bad faith action. Resolution of the ensuing inquiry into the employer's motives has been difficult to predict and demonstrates the imprecision of the standards thus far formulated. (See, e.g., Gould, Stemming the Tide, supra, 13 Emp.Rel.L.J. at pp. 405-407.) This situation undermines the statutory mandate that neither compensatory tort damages nor exemplary damages are available in an action arising from the breach of a contract obligation. (Civ. Code, §§ 3333, 3294.) fn. 38 Adoption of tests such as those formulated by the Court of Appeal would result in the anomalous result that henceforth the implied covenant in an employment contract would enjoy protection far greater than that afforded to express and implied-in-fact promises, the breach of which gives rise to an action for contract damages only. fn. 39 [47 Cal. 3d 699]

The Koehrer court recognized the problem of distinguishing between breaches of the contract and breaches of obligations imposed by law. fn. 40 It failed, however, to recognize that in traditional contract law, the motive of the breaching party generally has no bearing on the scope of damages that the injured party may recover for the breach of the implied covenant; the remedies are limited to contract damages. fn. 41 Thus, recitation of the parameters of the implied covenant alone is unsatisfactory. If the covenant is implied in every contract, but its breach does not in every contract give rise to tort damages, attempts to define when tort damages are appropriate simply by interjecting a requirement of "bad faith" do nothing to limit the potential reach of tort remedies or to differentiate between those cases properly and traditionally compensable by contract damages and those in which tort damages should flow. Virtually any firing (indeed any breach of a contract term in any context) could provide the basis for a pleading alleging the discharge was in bad faith under the cited standards.

Finally, and of primary significance, we believe that focus on available contract remedies offers the most appropriate method of expanding available relief for wrongful terminations. The expansion of tort remedies in the employment context has potentially enormous consequences for the stability of the business community.

We are not unmindful of the legitimate concerns of employees who fear arbitrary and improper discharges that may have a devastating effect on their economic and social status. Nor are we unaware of or unsympathetic to claims that contract remedies for breaches of contract are insufficient because they do not fully compensate due to their failure to include attorney fees and their restrictions on foreseeable damages. These defects, however, exist generally in contract situations. As discussed above, the variety of possible courses to remedy the problem is well demonstrated in the literature and includes increased contract damages, provision for award of attorney fees, establishment of arbitration or other speedier and less expensive [47 Cal. 3d 700] dispute resolution, or the tort remedies (the scope of which is also subject to dispute) sought by plaintiff here.

The diversity of possible solutions demonstrates the confusion that occurs when we look outside the realm of contract law in attempting to fashion remedies for a breach of a contract provision. As noted, numerous legislative provisions have imposed obligations on parties to contracts which vindicate significant social policies extraneous to the contract itself. As Justice Kaus observed in his concurring and dissenting opinion in White v. Western Title Ins. Co. (1985) 40 Cal. 3d 870, 901 [221 Cal. Rptr. 509, 710 P.2d 309], "our experience in Seaman's surely tells us that there are real problems in applying the substitute remedy of a tort recovery -- with or without punitive damages -- outside the insurance area. In other words, I believe that under all the circumstances, the problem is one for the Legislature ...."

Conclusion

Plaintiff may proceed with his cause of action alleging a breach of an implied-in-fact contract promise to discharge him only for good cause; his claim is not barred by the statute of frauds. His cause of action for a breach of public policy pursuant to Tameny was properly dismissed because the facts alleged, even if proven, would not establish a discharge in violation of public policy. Finally, as to his cause of action for tortious breach of the implied covenant of good faith and fair dealing, we hold that tort remedies are not available for breach of the implied covenant in an employment contract to employees who allege they have been discharged in violation of the covenant. fn. 42

Accordingly, that portion of the judgment of the Court of Appeal affirming the dismissal of plaintiff's causes of action alleging a discharge in breach of public policy and a tortious breach of the implied covenant of good faith and fair dealing is affirmed. That portion of the judgment of the Court of Appeal affirming the dismissal of the cause of action alleging an implied-in-fact contract not to discharge except for good cause is reversed, and the case is remanded for action consistent with the views expressed herein. fn. 43

Panelli, J., Arguelles, J., Eagleson, J., concurred. [47 Cal. 3d 701]

BROUSSARD, J.

I concur in part I of the majority opinion, which holds that plaintiff has not stated a cause of action for discharge in violation of public policy. I join fully in part II of the majority opinion, which upholds plaintiff's cause of action for breach of contract, but add a note exploring the question of the damages recoverable in such an action. I respectfully dissent to part III of the majority opinion. Although written in conservative tones of deference to legislative action, it is in fact a radical attempt to rewrite California law in a manner which, as the majority themselves acknowledge, will leave the wrongfully discharged worker without an adequate remedy. The majority opinion does not preserve the status quo, leaving it to the Legislature to adopt innovative solutions. It uproots the status quo, leaving it to the Legislature to remedy the problems the opinion creates.

I.

Under the majority opinion, employees will no longer have a tort cause of action for bad faith discharge, but, absent some violation of public policy, can sue only in contract. The majority acknowledge that traditional contract damages may provide inadequate compensation (see ante, pp. 699-700). They recognize that a considerable number of commentators have suggested that the remedy for widely perceived inequities in the contract-damages realm may lie in an expansion of the nature of damages that may properly be recovered within a breach of contract action. (See, e.g., Traynor, Bad Faith Breach of a Commercial Contract: A Comment on the Seaman's Case (Cal. State Bar, Fall 1984) 8 Bus. L. News 1 (hereafter Traynor); Putz & Klippen, Commercial Bad Faith: Attorney Fees -- Not Tort Liability -- Is the Remedy for "Stonewalling" (1987) 21 U.S.F. L.Rev. 419.) They conclude, however, that we should not resolve questions concerning the measure of contract damages in the present case. (Ante, p. 682, fn. 24.) I agree that we should not resolve the issue in the present case. But I think it appropriate in a concurring and dissenting opinion to suggest a line of reasoning which may prove fruitful.

As a general rule, damages recoverable in contract actions have been limited to those within the contemplation of the parties at the time of the contract. (See Hadley v. Baxendale (1854) 9 Ex. 341 [156 Eng.Rep. 145]; Hunt Bros. Co. v. San Lorenzo Water Co. (1906) 150 Cal. 51, 56 [87 P. 1093].) Since the ordinary commercial contract does not contemplate damages for mental or emotional distress, this rule has led to the maxim that damages for mental suffering are generally not recoverable in an action for breach of contract. (Westwater v. Grace Church (1903) 140 Cal. 339, 342 [73 P. 1055]; Allen v. Jones (1980) 104 Cal. App. 3d 207, 211 [163 Cal. Rptr. 445] and cases there cited.) But it has always been clear that this maxim is [47 Cal. 3d 702] subject to the obvious exception: if the contracting parties did contemplate that breach of the contract would cause emotional distress, damages for that injury are recoverable.

In Allen v. Jones, supra, 104 Cal. App. 3d 207, for example, defendant funeral home breached a contract to transport the cremated remains of plaintiff's brother. Because it was reasonably foreseeable that such breach would cause mental anguish to plaintiff, the court upheld a cause of action for mental distress. (Id. at pp. 214-215.) In Ross v. Forest Lawn Memorial Park (1984) 153 Cal. App. 3d 988, 995 [203 Cal. Rptr. 468, 42 A.L.R.4th 1049], the court upheld a claim for mental distress occasioned by breach of a contract to protect a grave from vandalism. In Wynn v. Monterey Club (1980) 111 Cal. App. 3d 789, 799-800 [168 Cal. Rptr. 878], plaintiff claimed that defendants' breach of a contract to bar plaintiff's wife from their gaming tables led to the breakup of his marriage; the court found a triable issue of fact whether damages for mental suffering were reasonably foreseeable and within the contemplation of the contracting parties. Windeler v. Scheers Jewelers (1970) 8 Cal. App. 3d 844 [88 Cal. Rptr. 39] upheld recovery for emotional distress when defendant, in breach of a contract of bailment, lost jewelry of great sentimental value.

These precedents have not yet been applied to wrongful discharge. But a review of the facts of the Court of Appeal cases overruled by the majority in part III of their opinion, and similar cases pending before this court, makes it clear that in many cases the employer is aware at the time of the contract that bad faith discharge will create great mental and emotional distress. In such cases, the application of existing precedent could serve to provide some redress for that injury. fn. 1

II.

The majority's discussion of the cause of action for wrongful discharge transposes the positions of plaintiffs and defendants, asserting, incorrectly, [47 Cal. 3d 703] that plaintiffs seek to change established law. For example, the majority introduce their analysis with the statement that "where an extension of tort remedies is sought for a duty whose breach previously has been compensible by contractual remedies, it is helpful to consider certain principles relevant to contract law." (Ante, p. 683, italics added.) After considering these principles, the majority conclude that "focus on available contract remedies offers the most appropriate method of expanding available remedies for wrongful termination." (Ante, p. 699, italics added.) But this case is not about extending or expanding remedies for wrongful discharge. All plaintiffs seek, and all the dissenters seek, is to retain existing remedies. It is the defendants, and the majority, who seek a radical contraction in existing remedies.

I maintain that we should retain the well-recognized tort cause of action for bad faith discharge. To demonstrate the point, I propose to show (1) that a tort cause of action for bad faith discharge is an established feature of California common law, (2) that the analogy between the insurance cases, in which a tort cause of action has long been recognized, justifies tort recovery for bad faith discharge; (3) that the existance of a cause of action in contract for discharge in breach of contract does not exclude a tort action for bad faith; and (4) that it is fundamentally illogical to abolish a tort cause of action on the ground that radical change in existing remedies should be left to legislative action.

1. A tort remedy for bad faith discharge is well established in California law.

Prior to Tameny v. Atlantic Richfield Co. (1980) 27 Cal. 3d 167 [164 Cal. Rptr. 839, 610 P.2d 1330, 9 A.L.R.4th 314], California decisions had recognized a tort action for bad faith breach of insurance contracts. (See, e.g., Comunale v. Traders & General Ins. Co. (1958) 50 Cal. 2d 654, 658 [328 P.2d 198, 68 A.L.R.2d 883]; Gruenberg v. Aetna Ins. Co. (1973) 9 Cal. 3d 566, 578 [108 Cal. Rptr. 480, 510 P.2d 1032].) In upholding a tort cause of action for wrongful discharge in violation of public policy, Tameny noted an alternative theory -- breach of the employer's duty of good faith and fair dealing. "[P]ast California cases," we said, citing the insurance cases, "have held that a breach of this implied-at-law covenant sounds in tort as well as in contract." (Tameny, supra, 27 Cal. 3d 167, 179, fn. 12.) In Tameny, however, we found it unnecessary to decide whether tort recovery would be available under that theory.

That issue was first decided in Cleary v. American Airlines (1980) 111 Cal. App. 3d 443 [168 Cal. Rptr. 722]. Plaintiff in Cleary pled that the employer, despite a written policy to the contrary, arbitrarily discharged him [47 Cal. 3d 704] after 18 years of satisfactory service. The Court of Appeal unanimously ruled that "[s]hould plaintiff sustain his burden of proof, he will have established a cause of action for wrongful discharge that sounds in both contract and in tort." (P. 456, italics added.)

The next case, Crosier v. United Parcel Service, Inc. (1983) 150 Cal. App. 3d 1132 [198 Cal. Rptr. 361], endorsed Cleary, which it described as based on "present economic realities and the reasonable expectations of the parties" (p. 1137), but concluded that since the company was enforcing a general rule known to the employee, the discharge was not in bad faith. A subsequent case, Shapiro v. Wells Fargo Realty Advisors, (1984) 152 Cal. App. 3d 467 [199 Cal. Rptr. 613], similarly applied the reasoning of Cleary, but found plaintiff's allegations insufficient to show bad faith.

In Seaman's Direct Buying Service, Inc. v. Standard Oil Co. (1984) 36 Cal. 3d 752 [206 Cal. Rptr. 354, 686 P.2d 1158], we concluded a tort action was available for breach of a commercial contract only when the breaching party denied in bad faith the existence of the contract. (P. 769.) We noted that tort remedies had a broader scope in insurance cases because of the special relationship between insurer and insured, and added that "no doubt there are other relationships with similar characteristics and deserving of similar legal treatment." (P. 769.) The footnote to that sentence noted that in Tameny v. Atlantic Richfield Co., supra, 27 Cal. 3d 167, 179, footnote 12, "this court intimated that breach of the covenant of good faith and fair dealing in the employment relationship might give rise to tort remedies. That relationship has some of the same characteristics as the relationship between insurer and insured." (36 Cal.3d at p. 769, fn. 6.) Coming after published decisions in Cleary v. American Airlines, Inc., supra, 111 Cal. App. 3d 443, Shapiro v. Wells Fargo Realty Advisors, supra, 152 Cal. App. 3d 467, and Crosier v. United Parcel Service, supra, 150 Cal. App. 3d 1132, this language signaled the court's approval of a tort remedy for bad faith discharge. fn. 2

Shortly after Seaman's was filed, the court in Rulon-Miller v. International Business Machines Corp. (1984) 162 Cal. App. 3d 241 [208 Cal. Rptr. 524], considered whether defendant could fire an employee for "conflict of interest" because she was dating a fellow employee. The case was submitted to the jury under instructions based on Cleary v. American Airlines, Inc., [47 Cal. 3d 705] supra, 111 Cal. App. 3d 443. The jury awarded tort damages, finding the asserted conflict of interest a mere pretext and the discharge in bad faith. The Court of Appeal affirmed; we denied a hearing.

The next two cases, Wallis v. Superior Court (1984) 160 Cal. App. 3d 1109 [207 Cal. Rptr. 123], and Wayte v. Rollins International, Inc. (1985) 169 Cal. App. 3d 1 [215 Cal. Rptr. 59], did not involve discharge. (Wallis concerned bad faith refusal of an employer to pay retirement benefits; Wayte concerned bad faith refusal of an employer acting as an insurer to pay medical benefits.) Both cases, however, affirmed the analogy of the employment relationship to insurance. As stated in Wallis, "the characteristics of the insurance contract which give rise to an action sounding in tort are also present in most employer-employee relationships." (160 Cal. App. 3d 1109, 1116, fn. 2.)

Khanna v. Microdata Crop. (1985) 170 Cal. App. 3d 250 [215 Cal. Rptr. 860] is recognized by the majority as one of the two leading cases on bad faith discharge. In that case the employer refused to acknowledge a contract to pay plaintiff a commission, then fired the employee when he sued to enforce the contract. The opinion reviewed Cleary, supra, 111 Cal. App. 3d 443, and other California decisions and concluded that a tort cause of action is stated whenever the employer engages in "'bad faith action extraneous to the contract, combined with the obligor's intent to frustrate the [employee's] enjoyment of contract rights.'" (Khanna, supra, at p. 262, quoting Shapiro v. Wells Fargo Realty Advisors, supra, 152 Cal.App.3d at pp. 478-479.) Gray v. Superior Court (1986) 181 Cal. App. 3d 813 [226 Cal. Rptr. 570], relied on Khanna to find that a worker fired on the basis of a false performance report could state a cause of action in tort.

The other leading case, Koehrer v. Superior Court (1986) 181 Cal. App. 3d 1155 [226 Cal. Rptr. 820], was a decision by Justice Kaufman, now on this court. It is well summarized in his dissenting opinion here. As noted in Koehrer, a tort cause of action for bad faith discharge may arise if the employer asserts the existence of good cause for discharge without probable cause and in bad faith. (P. 1171.)

Finally, the Court of Appeal in the present case adopted the most limited view of that cause of action of any of the California decisions. It claimed that to state a tort cause of action, the plaintiff must allege facts comparable to those in Cleary -- 18 years longevity, and the employer's violation of specific employment guidelines. But, for cases within those facts, the Court of Appeal would permit a tort action.

In sum, there are eight unanimous Court of Appeal decisions permitting a tort action for bad faith discharge, plus dictum approving such an action [47 Cal. 3d 706] in cases here and in the Court of Appeal. Thus it is not surprising that when the Ninth Circuit considered the matter in a case arising under California law, it had no doubt that such a cause of action existed. (See Huber v. Standard Ins. Co. (9th Cir. 1988) 841 F.2d 980.) The only unsettled question, from the viewpoint of the federal judges, was whether the tort cause of action was limited to cases comparable to Cleary, as the Court of Appeal held in the present case. The Ninth Circuit correctly concluded that it was not so limited, citing Koehrer v. Superior Court, supra, 181 Cal. App. 3d 1155-1169 and Khanna v. Microdata Corp., supra, 170 Cal. App. 3d 250, 262 as declarative of California law.

Many commentators have written summaries or analyses of California law in this area. (See Cal. Wrongful Employment Termination Practice (Cont.Ed.Bar 1987) § 2.42; Kornblum et al., Cal.Practice Guide: Bad Faith (1986) § 12.13; McCarthy, Punitive Damages in Wrongful Discharge Cases (1985) § 2.2; Brandon, From Tameny to Foley. Time for Constitutional Limitations on California's Employment at Will Doctrine (1988) 15 Hastings Const. L.Q. 359, 371-372 (hereafter Brandon); Brody, Wrongful Termination as Labor Law (1988) 17 Sw. U.L. Rev. 434, 442.) All recent writings consider the existence of a tort cause of action for bad faith discharge an established part of the California common law.

Such unanimous agreement among justices and commentators generates reliance. fn. 3 Employers have revised personnel policies and purchased insurance policies. Insurers have calculated and collected premiums. Attorneys have been hired and trained, even entire law firms have been established. Litigants have filed suits, accepted settlement offers, rejected other offers, gone to trial, and appealed. Hundreds of cases are proceeding before the trial courts in which both parties have based their strategy on the assumption that a tort action exists. Many others pending in the Court of Appeal await our decision. There are 10 or so such cases pending before this court. fn. 4

The majority do not deny the existence of prior precedent, or the consequent reliance thereon, but point out that there has been no decision by this court declaring a cause of action for bad faith discharge. They assert that "[i]f we were to follow the dissent's urging that we should therefore leave this area of law untouched, we would be abdicating our role." (Ante, p. 689, [47 Cal. 3d 707] fn. 28.) They mistake my purpose. I do not claim that this court should never repudiate prior precedent and revise California common law, but only that when it considers doing so it should recognize that the doctrine of stare decisis and judicial reluctance to penalize justified reliance on precedent weigh heavily against such a decision. The majority here in fact radically restructure California law by abolishing an established cause of action, yet they write as if theirs is nothing more than a conservative decision declining to extend remedies into uncharted realms. I cannot join them in this retrogressive decision.

2. Analogy to the cases upholding a tort cause of action for bad faith breach of a contract of insurance justifies a tort action for bad faith discharge of an employee.

The majority deride the prior cases for their uncritical incorporation of the insurance model into the employment context without considering the significant differences between the insurer-insured and employer-employee relationships. (Ante, p. 689.) But when we consider the differences noted by the majority, we find that they are not significant at all.

The majority find one fundamental difference between insurance and employment relationships: "[i]f an insurer pays a claim, it diminishes its fiscal resources ... [while] as a general rule it is to the employer's economic benefit to retain good employees." (Ante, at p. 693.) But their comparison is not between insurers and employers, but between short-sighted insurers and far-sighted employers. In the short run, the insurer saves money by not paying claims, and the employer by not paying wages. (If the work cannot be deferred, he can hire less experienced but cheaper help.) In the long run, an insurer that never paid claims would be out of business, and an employer that always fired experienced help would not be much better off. Thus if we examine insurers and employers with the same lens, the difference the majority find fundamental simply disappears.

But the majority's analysis leaves a lingering trace, for it betrays their misunderstanding of the problem. We need not be concerned about insurers that never pay claims or employers that fire all experienced help -- the marketplace will take care of them. The concern is with the insurer or employer that acts arbitrarily some of the time -- and can get away with it unless threatened with damages that, unlike traditional contract damages, exceed the short-term profit. fn. 5 [47 Cal. 3d 708]

The majority also point to some nonfundamental distinctions between the insurer-insured and the employer-employee relationships. They argue that the discharged employee may be able to mitigate damages while the insured generally cannot. But as we all know, in many cases the discharged worker cannot mitigate damages. As Justice Kaufman asks, "What market is there for the factory worker laid-off after 25 years of labor in the same plant, or for the middle-aged executive fired after 25 years with the same firm?" (Con. and dis. opn., ante, at p. 718.) The ability of some persons to mitigate damages is no reason to deny a cause of action to those unable to mitigate them.

It is next suggested that the employer, unlike the insurer, is not performing a "public service." I fail to understand the significance of the statement. Employment is even more important to the community than insurance; most people value their jobs more than their insurance policies. The public interest in deterring arbitrary breach of employment contracts is, I suggest, at least equal to that in deterring arbitrary breach of insurance contracts.

Finally, the majority reject the idea that an employee is like an insured because both contract for financial security. A business, they point out, may also seek financial security. They put the case of a business contracting to secure a reliable source of supply. But what emerges from the majority's analysis is three propositions: a) that insureds generally buy insurance policies for financial security; (b) that employees generally seek financial security in their employment; (c) that businesses occasionally contract for financial security. These propositions should lead the majority to conclude that the employment contract is more analogous to an insurance contract than to a commercial contract.

The majority are focusing upon the exceptions, not upon the general rule. If we must argue analogies, the question is not whether the employment contract differs from an insurance contract in one particular respect, or resembles a commercial contract in another. It is whether, as a whole, the [47 Cal. 3d 709] contract of employment more closely resembles an insurance contract or an ordinary commercial contract. The answer is clear. The principal reason we permit tort damages for breach of the covenant of good faith and fair dealing in an insurance contract is that persons do not generally purchase insurance to obtain a commercial advantage, but to secure the peace of mind and security it will provide in protecting against accidental loss. (See Crisci v. Security Ins. Co. (1967) 66 Cal. 2d 425, 434 [58 Cal. Rptr. 13, 426 P.2d 173].) That reason applies equally to the employer-employee relationship. A man or a woman usually does not enter into employment solely for the money; a job is status, reputation, a way of defining one's self-worth and worth in the community. It is also essential to financial security, offering assurance of future income needed to repay present debts and meet future obligations. Without a secure job a worker frequently cannot obtain a retirement pension, and often lacks access to affordable medical insurance. In short, "in a modern industrialized economy employment is central to one's existence and dignity." (Gould, The Idea of the Job as Property in Contemporary America: The Legal and Collective Bargaining Framework, 1986 B.Y.U. L. Rev. 885, 892.) fn. 6

Because workers value their jobs as more than merely a source of money, contract damages, if limited to loss of income, are inadequate. (See K Mart Corp. v. Ponsock, supra, 732 P.2d 1364, 1371; Comment, Reconstructing Breach of the Implied Covenant of Good Faith and Fair Dealing as a Tort (1985) 73 Cal.L.Rev. 1291, 1330; cf. Traynor, op. cit. supra, 8 Bus. L. News 1, 13.) Again the analogy to the insurance cases is close. Explaining the basis for tort damages in insurance cases, Wallis v. Superior Court, supra, 160 Cal. App. 3d 1109, 1118, said that "[m]oney damages paid pursuant to a judgment years after ... do not remedy the harm suffered ..., namely the immediate inability to support oneself and its attendant horrors" -- language which applies equally to a suit for wrongful discharge. As summarized in Miller & Estes, Recent Judicial Limitations On the Right to Discharge: A California Trilogy (1982) 16 U.C. Davis L.Rev. 65, 90-91, insureds and employees both depend on the contracts "for their security, well-being, and peace of mind. If insurance companies or employers act in bad faith, the consequences can be very severe, indeed much greater than those that result from a breach of contract."

In contrast, commercial contracts, generally speaking, are negotiated between parties of more nearly equal bargaining strength, and are entered [47 Cal. 3d 710] into for purpose of profit. Breach entails only lost profits, and often a market exists in which the damaged party can cover its loss. I conclude that past decisions were justified in analogizing the relationship between employer and employee to that between insurer and insured, and in distinguishing both from commercial contracts for the sale of goods and services.

3. The existence of a contract action for discharge in breach of contract does not exclude a tort action for bad faith.

The majority also assert that the prior cases have not carefully considered the fundamental policies underlying the development of tort and contract law. (Ante, at p. 689.) Their argument, in essence, is that the covenant of good faith and fair dealing is simply one provision in the contract. When the employer acts in "bad faith," they argue, he breaches the contract, and the appropriate remedy is an action for damages recoverable for breach of contract.

There are several objections to this reasoning. First, as pointed out by Justice Kaufman, the covenant is a duty imposed by law, not one arising from the terms of the contract. (Conc. & dis. opn. of Kaufman, J., ante, at p. 716).) Second, the majority's reasoning was previously advanced in opposition to tort recovery in the insurance cases, and has there been rejected by this court. Finally, the majority mistake the distinction between a contract action for wrongful discharge and a tort action for bad faith discharge.

A suit in contract for wrongful discharge is not based directly upon the covenant of good faith and fair dealing, but upon some other provision limiting the right of the employer to discharge at will. fn. 7 A provision may prohibit discharge without "good cause," in which the employer's good faith may be relevant in deciding whether good cause exists. (Pugh v. See's Candies, Inc. (1981) 116 Cal. App. 3d 311, 330 [171 Cal. Rptr. 917].) The provision may, instead, establish some other limitation on the employer's power to discharge the employee. It may, for example, prohibit the discharge of the worker before January 1, or so long as he produces 400 widgets per month, or without an opportunity for a grievance hearing -- in which examples good faith would not be relevant. The point is that all such actions rest not on proof of the employer's bad faith, but of his breach of some contractual provision apart from the covenant.

A tort action for bad faith discharge also requires that the discharge be wrongful -- that is, in breach of contract. But once that prerequisite is satisfied, [47 Cal. 3d 711] it focuses not upon the employee's right to enforce a particular contractual provision, but upon society's right to deter and demand redress for arbitrary or malicious conduct which inflicts harm upon one of its members. This is the proper and traditional function of tort law.

The majority attack a tort remedy because it is different than a contract remedy, as if tort law itself served no purpose. They argue, for example, that abolition of a tort remedy for bad faith discharge will enhance predictability of damages. fn. 8 The unpredictability of damages in a tort action for bad faith discharge is the consequence of allowing recovery for emotional distress. Damages in other torts permitting recovery for emotional distress, such as negligence, products liability, malpractice, intentional infliction of emotional distress, etc., are equally unpredictable. I see no reason why predictability is more important to employers in connection with wrongful discharge actions than, for example, in actions for injuries caused by defective products. Of course, one can enhance predictability by denying recovery for injuries suffered, but this is not a trade-off courts have generally been willing to make. It is a decision which, by the majority's own logic, is better left to the Legislature.

The majority then argue that it would be difficult if not impossible to formulate a rule to confine tort relief to "deserving" cases, apparently because the concept of "good faith" is subjective. fn. 9 In fact, a suitable test is simple to describe: an employer acts in bad faith in discharging an employee if and only if he does not believe he has a legal right to discharge the employee. fn. 10 The majority assert that under similar tests employed by the courts "an ordinary contract breach might give rise to a bad faith action" (ante, p. 698) and that a requirement of bad faith "do[es] nothing ... to differentiate between those cases properly and traditionally compensable by contract damages and those in which tort damages should flow" (ante, p. 699). But these assertions are obviously mistaken. A breach of contract does [47 Cal. 3d 712] not require bad faith. The distinction between contract and tort is between a discharge done in good faith, where the employer believes he has a legal right to discharge the worker, and deliberate, arbitrary violation of the employee's rights. Indeed, the majority acknowledge in a footnote the effectiveness of a similar test in limiting tort recovery, and complain only that it does not "serve to limit initiation and prosecution of litigation based on almost any discharge." (Ante, p. 697, fn. 35.) I see no reason why the defense remedies of demurrer and summary judgment will not prove as effective here as in any other tort action: presumably if plaintiff cannot prove bad faith he will not assert it in his complaint; if he does he will lose on summary judgment. fn. 11

The majority particularly attack Koehrer v. Superior Court, supra, 181 Cal. App. 3d 1155, on the ground that "it failed, however, to recognize that in traditional contract law, the motive of the breaching party generally has no bearing on the scope of damages that the injured party may recover for breach of the implied covenant." (Ante, p. 699, italics in original.) The argument seems misplaced in the majority opinion; it belongs in the dissent. The fact that traditional contract law draws no distinction between good faith and bad faith, between innocent breach and malicious breach, is a good reason for looking to tort law where such distinctions are recognized. Indeed, the majority argument on this point, and, one might say, on the entire issue of tort damages, makes no sense unless the majority believe that there should be no distinction between innocent and malicious breach -- that the employer who maliciously and arbitrarily fires a worker knowing that he has no right to do so should pay no more in damages than the employer who believed in good faith that he had a right to fire the worker -- and in particular that the bad faith employer should not pay for the suffering he knowingly and deliberately caused. That is not a belief which I share.

4. It is fundamentally illogical for the majority to abolish an established tort remedy for bad faith discharge and then to argue that radical change in existing remedies is best left to the Legislature.

The majority in their concluding words recognize the need to provide wrongfully discharged workers with an adequate remedy. They also observe that traditional contract remedies may be inadequate. They assert, however, [47 Cal. 3d 713] that an action in tort is not necessarily the answer. Commentators, they say, have pointed out that a tort remedy has social and economic implications which could be more easily studied and weighed by a legislature than by the courts. fn. 12 Some have suggested alternative remedies, such as arbitration, which could be enacted by legislation but not by judicial decision. Thus, according to the majority, the choice between tort remedies and alternative remedies is one best made by the Legislature, not by the courts. fn. 13 One reading this argument would expect that the majority opinion would be limited to clarifying existing law relating to the availability of a tort remedy and that any drastic change in existing law, such as the abolition of the tort remedy or its replacement by some other remedy, would be left to the Legislature.

But the majority do not clarify existing law. They repudiate it. They reject the guidance of the only cases by this court to discuss tort actions for wrongful discharge. They reverse the Court of Appeal decision in this case. They overrule seven other unanimous Court of Appeal decisions -- every Court of Appeal decision to decide that issue. fn. 14 They will cause reversals in [47 Cal. 3d 714] approximately 10 other cases in which review has been granted and held for this decision, since every one of those cases recognized a tort cause of action. Having thus swept the table clean of California precedent, fn. 15 they adopt a rule limiting workers discharged in bad faith to contract damages -- a rule rejected by most of the commentators on whom the majority rely, and recognized as inadequate by the majority themselves.

Only after completely rewriting California law on the subject do the majority leave the matter to the Legislature. But two things have changed. First, the Legislature will face a problem -- the inadequacy of damages in actions for bad faith discharge -- which did not exist before. Second, the burden of seeking legislative change, which was previously on employers and insurers, two well-organized and financed groups, is now on the unorganized worker.

In fact, the Legislature has already considered this matter. In the 1985-1986 legislative session Senator Greene and Assemblyman McAllister both introduced bills which would have abolished a tort remedy for bad faith discharge, and provided for arbitration of contract claims. (See Sen. Bill No. 1348 and Assem. Bill No. 2800, summarized in Brandon, op. cit. supra, 15 Hastings Const. L.Q. 359, 373-374.) Neither bill was enacted. One cannot infer too much from the Legislature's failure to enact a bill, but it seems safe to say that the Legislature was not contemplating abolishing a tort remedy without a suitable replacement. [47 Cal. 3d 715]

5. Conclusion.

The majority's action in abolishing the tort remedy for bad faith discharge comes soon after its decision in Moradi-Shalal v. Fireman's Fund Ins. Companies, supra, 46 Cal. 3d 287, which abolished the tort cause of action for bad faith settlement practices by insurers. In both cases the court has reached out to overturn precedent and to abolish a cause of action recognized by California law. The decisions abolish protections previously accorded consumers and workers; they extend protection to insurers and employers, giving them tort immunity not for innocent error but for bad faith breach of duty. In both cases, moreover, the California Legislature has considered abolishing the cause of action in question but refrained from so doing; fn. 16 the decisions of this court grant immunities which the Legislature has declined to give, and shift to the unorganized consumer and worker the burden of seeking legislative change. It is the function of the common law "to protect the weak from the insults of the stronger." (3 Blackstone, Commentaries 3.) The majority's decision subverts that function.

I have considerable respect for the doctrine of judicial restraint, but that doctrine must run both ways. Judicial restraint should not only restrain the court from creating new remedies, it should also restrain the court from dismantling existing ones.

KAUFMAN, J.

I concur in parts I and II of the majority opinion. However, I respectfully dissent from part III dealing with breach of the implied duty of good faith and fair dealing. Contrary to the majority, I would hold that such a breach in the context of employment termination may give rise to tort remedies. The reasons which impel me to this conclusion are set forth below.

Breach of the Implied Duty of Good Faith and Fair Dealing

Thirty years ago, in Comunale v. Traders & General Ins. Co. (1958) 50 Cal. 2d 654 [328 P.2d 198, 68 A.L.R.2d 883], this court first recognized that breach of the implied duty of good faith and fair dealing may give rise to a cause of action sounding in tort. I would not have thought, after these many years, that it was still necessary to defend and explain this basic principle. In purporting to trace its origins, however, the majority fundamentally misstates the nature of the tort, and thereby subverts the powerful impetus for its extension to the area of employment termination. A brief summary of familiar principles, therefore, may be useful. [47 Cal. 3d 716]

In attempting to emphasize its contractual origins, the majority characterize the covenant of good faith and fair dealing fn. 1 as "a contract term" (maj. opn. at p. 684) "aimed at making effective the agreement's promises." (Maj. opn. at p. 683.) That characterization is simply incorrect under the decisions of this court and the authorities on which they relied. It is true that the law implies in every contract a duty of good faith and fair dealing. (Seaman's Direct Buying Service, Inc. v. Standard Oil Co. (1984) 36 Cal. 3d 752, 768 [206 Cal. Rptr. 354, 686 P.2d 1158]; Egan v. Mutual of Omaha Ins. Co. (1979) 24 Cal. 3d 809, 818 [169 Cal. Rptr. 691, 620 P.2d 141].) The duty to deal fairly and in good faith with the other party to a contract, however, "is a duty imposed by law, not one arising from the terms of the contract itself. In other words, this duty of dealing fairly and in good faith is nonconsensual in origin rather than consensual." (Richardson v. Employers Liab. Assur. Corp. (1972) 25 Cal. App. 3d 232, 239 [102 Cal. Rptr. 547], italics added, quoted with approval in Gruenberg v. Aetna Ins. Co. (1973) 9 Cal. 3d 566, 574 [108 Cal. Rptr. 480, 510 P.2d 1032]; accord Tameny v. Atlantic Richfield Co. (1980) 27 Cal. 3d 167, 175-176 [164 Cal. Rptr. 839, 610 P.2d 1330, 9 A.L.R.4th 314].) While the nature of the obligations imposed by this duty is dependent upon the nature and purpose of the contract and the expectations of the parties, these obligations are not consensual, not agreed to in the contract; they are imposed by law and thus reflect the normative values of society as a whole. (Gruenberg v. Aetna Ins. Co., supra, 9 Cal.3d at pp. 573-575; Koehrer v. Superior Court (1986) 181 Cal. App. 3d 1155, 1169 [226 Cal. Rptr. 820].) The interest which the duty of good faith and fair dealing is designed to preserve and protect is essentially not the parties' interest in having their promises performed, but society's interest in protecting its members from harm on account of nonconsensual conduct. (Gruenberg v. Aetna Ins. Co., supra, 9 Cal.3d at pp. 573-574; Koehrer v. Superior Court, supra, 181 Cal.App.3d at pp. 1169-1171.)

Because tort actions enforce "duties of conduct ... imposed by law, and are based primarily upon social policy, and not necessarily upon the will or intention of the parties ...." (Tameny v. Atlantic Richfield Co., supra, 27 Cal.3d at p. 176, quoting Prosser, Law of Torts (4th ed. 1971) p. 613), it was quite natural that courts would eventually approve the extension of tort remedies, in appropriate circumstances, to violations of the duty of good faith and fair dealing. (See Note, supra, 73 Cal.L.Rev. at p. 1307.) Indeed, [47 Cal. 3d 717] this court was among the first to recognize that the nature of the obligations, the purposes of the contract and the expectations of the parties all combine to impose a heightened duty upon insurers. As we explained in Egan v. Mutual of Omaha Ins. Co., supra, 24 Cal. 3d 809, "The insured in a contract like the one before us does not seek to obtain a commercial advantage by purchasing the policy -- rather, he seeks protection against calamity .... [T]he major motivation for obtaining disability insurance is to provide funds during periods when the ordinary source of the insured's income -- his earnings -- has stopped. The purchase of such insurance provides peace of mind and security in the event the insured is unable to work." (24 Cal.3d at p. 819.) We also observed that "the relationship of insurer and insured is inherently unbalanced; the adhesive nature of insurance contracts places the insurer in a superior bargaining position." (Id. at p. 820.)

In the classic tradition of the common law, which adapts functional principles from precedent as changing social and economic conditions require, a number of courts and commentators have distilled from our holdings in the insurance context a relatively narrow but serviceable "bad faith" doctrine for application in other areas: Breach of the duty of good faith and fair dealing may give rise to an action in tort where the contractual relation manifests elements similar to those which characterize the "special relationship" between insurer and insured, i.e., elements of public interest, adhesion, and financial dependency. (Wallis v. Superior Court (1984) 160 Cal. App. 3d 1109, 1117-1119 [207 Cal. Rptr. 123]; Louderback & Jurika, Standards for Limiting the Tort of Bad Faith Breach of Contact (1981) 16 U.S.F. L.Rev. 187, 220-226; accord, Seaman's Direct Buying Service, Inc. v. Standard Oil Co., supra, 36 Cal.3d at pp. 768-769.)

Captured by the force of such reasoning, and fueled by dicta in Tameny and Seaman's suggesting that the employment relationship uniquely satisfies these several criteria (Tameny v. Atlantic Richfield Co., supra, 27 Cal.3d at p. 179, fn. 12; Seaman's Direct Buying Service, Inc. v. Standard Oil Co., supra, 36 Cal.3d at p. 769, fn. 6), recent Court of Appeal decisions have unanimously recognized that willful and malicious discharge from employment may give rise to tort remedies. (See Koehrer v. Superior Court, supra, 181 Cal.App.3d at pp. 1167-1171, Gray v. Superior Court (1986) 181 Cal. App. 3d 813, 820-821 [226 Cal. Rptr. 570]; Khanna v. Microdata Corp. (1985) 170 Cal. App. 3d 250, 260-264 [215 Cal. Rptr. 860]; Shapiro v. Wells Fargo Realty Advisors (1984) 152 Cal. App. 3d 467, 477-479 [199 Cal. Rptr. 613]; Rulon-Miller v. International Business Machines Corp. (1984) 162 Cal. App. 3d 241, 251-253 [208 Cal. Rptr. 524].)

The majority is not unmindful of these numerous authorities which have concluded that the criteria which make the relationship between insurer [47 Cal. 3d 718] and insured suitable for tort remedies, apply with even greater force in the employment context. Indeed, the majority reviews the pertinent cases and authorities with considerable care. (Maj. opn. at pp. 685-692.) At the end of this lengthy prologue, however, the majority concludes that all of the arguments are deficient in comparative analysis, and proceeds to explain why it is "not convinced" that a relationship analogous to that between insurer and insured exists in the employment context. (Maj. opn. at p. 692.) That explanation, in its entirety, is as follows.

First, the majority asserts that a breach in the employment context "does not place the employee in the same economic dilemma that an insured faces" because the insured "cannot turn to the marketplace," while an employee presumably may "seek alternative employment." (Maj. opn. at p. 692.) Next, the majority argues that an employer, unlike an insurance company, does not sell economic "protection." (Maj. opn. at p. 692.) The majority also rejects the insurance analogy because an employee, unlike an insured, allegedly does not seek a "different kind of financial security than those entering a typical commercial contract." (Maj. opn. at p. 692.) Finally, the majority asserts that insurance and employment contracts differ "fundamental[ly]" because the insured's and insurer's interests are "financially at odds," while the employer's and employee's interests allegedly are "most frequently in alignment." (Maj. opn. at p. 693.)

Such conclusions, in my view, expose an unrealistic if not mythical conception of the employment relationship. They also reveal a misplaced reluctance to define the minimal standards of decency required to govern that relationship. The delineation of such standards is not, as the majority strongly implies, judicial legislation, but rather constitutes this court's fundamental obligation.

It is, at best, naive to believe that the availability of the "marketplace," or that a supposed "alignment of interests," renders the employment relationship less special or less subject to abuse than the relationship between insurer and insured. Indeed, I can think of no relationship in which one party, the employee, places more reliance upon the other, is more dependent upon the other, or is more vulnerable to abuse by the other, than the relationship between employer and employee. And, ironically, the relative imbalance of economic power between employer and employee tends to increase rather than diminish the longer that relationship continues. Whatever bargaining strength and marketability the employee may have at the moment of hiring, diminishes rapidly thereafter. Marketplace? What market is there for the factory worker laid off after 25 years of labor in the same plant, or for the middle-aged executive fired after 25 years with the same firm? [47 Cal. 3d 719]

Financial security? Can anyone seriously dispute that employment is generally sought, at least in part, for financial security and all that that implies: food on the table, shelter, clothing, medical care, education for one's children. Clearly, no action for breach of the covenant of good faith and fair dealing will lie unless it has first been proved that, expressly or by implication, the employer has given the employee a reasonable expectation of continued employment so long as the employee performs satisfactorily. (Koehrer v. Superior Court, supra, 181 Cal.App.3d at p. 1171.) And that expectation constitutes a far greater and graver security interest than any which inheres in the insurance context. Most of us can live without insurance. Few of us could live without a job.

Peace of mind? One's work obviously involves more than just earning a living. It defines for many people their identity, their sense of self-worth, their sense of belonging. The wrongful and malicious destruction of one's employment is far more certain to result in serious emotional distress than any wrongful denial of an insurance claim.

If everything this court has written concerning the relation between insurer and insured has any deeper meaning; if we have created a living principle based upon justice, reason and common sense and not merely a fixed, narrow and idiosyncratic rule of law, then we must acknowledge the irresistible logic and equity of extending that principle to the employment relationship. We can reasonably do no less.

Beyond, and perhaps underlying, its rejection of the special relationship analogy, the majority makes two additional arguments. The extension of tort remedies to the employment relation, the majority asserts, is a matter best left to the Legislature, for the decision involves "policy" choices which may "profoundly" affect social and commercial relations. (Maj. opn. at p. 694.) Additionally, the majority asserts that "bad faith" is a tort impervious to practical "delineation." (Maj. opn. at p. 698.) Thus, the majority raises the familiar "floodgate-of-litigation" specter to warn of "potentially enormous consequences for the stability of the business community." (Maj. opn. at p. 699.) These sorts of arguments have not fared particularly well with this court in the past, and this case presents no occasion for exception.

As to the contention that a recognition of tort remedies for breach of the duty of good faith and fair dealing is best left to the Legislature, the short answer was aptly summarized in People v. Pierce (1964) 61 Cal. 2d 879, 882 [40 Cal. Rptr. 845, 395 P.2d 893], "In effect the contention is a request that courts abdicate their responsibility for the upkeep of the common law. That upkeep it needs continuously, as this case demonstrates." [47 Cal. 3d 720]

The responsibility to which we referred in Pierce, supra, 61 Cal. 2d 879, arises from the unique genius of the common law system and the critical role which the courts play in that system. "In California as in other jurisdictions of Anglo-American heritage, the common law 'is not a codification of exact or inflexible rules for human conduct ... but is rather the embodiment of broad and comprehensive unwritten principles, inspired by natural reason and an innate sense of justice, and adopted by common consent for the regulation and government of the affairs of men. ...'

"'The inherent capacity of the common law for growth and change is its most significant feature. Its development has been determined by the social needs of the community which it serves. It is constantly expanding and developing in keeping with advancing civilization and the new conditions and progress of society ....'" (Rodriguez v. Bethlehem Steel Corp. (1974) 12 Cal. 3d 382, 393-394 [115 Cal. Rptr. 765, 525 P.2d 669].)

"This flexibility and capacity for growth and adaptation is the peculiar boast and excellence of the common law." (Hurtado v. California (1884) 110 U.S. 516, 530 [28 L. Ed. 232, 237, 4 S. Ct. 111].)

"But that vitality can flourish only so long as the courts remain alert to their obligation and opportunity to change the common law when reason and equity demand it .... Although the Legislature may of course speak to the subject, in the common law system the primary instruments of evolution are the courts, adjudicating on a regular basis the rich variety of individual cases brought before them." (Rodriguez v. Bethlehem Steel Corp., supra, 12 Cal.3d at p. 394.)

This court has adhered to these principles over the years, consistently rejecting claims that the reconsideration, expansion or abolition of settled common law rules should await action by the Legislature. In Muskopf v. Corning Hospital Dist. (1961) 55 Cal. 2d 211 [11 Cal. Rptr. 89, 359 P.2d 457], for example, we judicially abolished the doctrine of sovereign immunity from tort liability, overruling earlier decisions which had expressly held that "abrogation or restriction of this doctrine is primarily a legislative matter ...." (Vater v. County of Glenn (1958) 49 Cal. 2d 815, 820 [323 P.2d 85]; see also Talley v. Northern San Diego Hosp. Dist. (1953) 41 Cal. 2d 33, 41 [257 P.2d 22].) In so holding, we emphasized that the doctrine was "court made" (55 Cal.2d at p. 218), and finding no continued "rational basis" for it (id. at p. 216), we ended it. fn. 2 [47 Cal. 3d 721]

In Li v. Yellow Cab Co. (1975) 13 Cal. 3d 804 [119 Cal. Rptr. 858, 532 P.2d 1226, 78 A.L.R.3d 393], this court judicially abolished the doctrine of contributory negligence as a total bar to recovery and adopted a rule of comparative negligence in direct proportion to fault. We did so despite the contention that "widespread disagreement among both the commentators and the states as to which [system was] best" rendered the matter more suitable for legislative action. (Id. at p. 833 (dis. opn. of Clark, J.).) "[L]ogic, practical experience, and fundamental justice," we held, made judicial action imperative. (Id. at pp. 812-813.)

Over the express objection that fundamental tort reform should be left to the Legislature, we judicially abandoned the long-standing distinctions among business invitees, social guests and trespassers in determining the standard of liability for dangerous conditions on land (Rowland v. Christian (1968) 69 Cal. 2d 108, 121 [70 Cal. Rptr. 97, 443 P.2d 561, 32 A.L.R.3d 496]); recognized that a married person whose spouse is injured by the negligence of a third party has a cause of action for loss of consortium (Rodriguez v. Bethlehem Steel Corp., supra, 12 Cal.3d at pp. 389-404); and abrogated the rule of interspousal immunity for negligent torts. (Klein v. Klein (1962) 58 Cal. 2d 692, 697-699 [26 Cal. Rptr. 102, 376 P.2d 70].)

Every one of these landmark decisions required a difficult choice among competing social and economic policies. It is hardly novel or surprising, therefore, to be confronted with conflicting arguments over the relative economic efficiency or political wisdom of allowing tort remedies for wrongful and malicious employment termination, or to find that competing policies have been advanced and adopted elsewhere. What is both novel and distressing is to find such arguments being cited by the majority as a justification for judicial abstention. The imposition of a tort duty is not contingent upon a consensus of so-called experts. The courts are the custodians of the common law -- not the economists, or the legislators, or even the law professors. We abdicate that duty when we abjure decision of common law questions under the guise of "deference" to the political branches.

The second contention of the majority, that bad faith liability cannot adequately be delimited and therefore must be denied, has also been generally rejected by this court. As we explained in Dillon v. Legg (1968) 68 Cal. 2d 728 [69 Cal. Rptr. 72, 441 P.2d 912], "The 'contention that a rule permitting the maintenance of the action would be impractical to administer ... is but an argument that the courts are incapable of performing their appointed tasks, a premise which has frequently been rejected.'" (Id. at [47 Cal. 3d 722] p. 736.) The unreasoning fear that we cannot successfully adjudicate future cases of bad faith termination, pursuant to the guidelines already formulated and applied by the Courts of Appeal, should not be permitted to bar recovery in an otherwise meritorious case.

Contrary to the implication of the majority, the standard set forth in Koehrer v. Superior Court, supra, 181 Cal. App. 3d 1155, is fully adequate to distinguish between simple breach of contract for discharge without cause, and breach of the implied duty of good faith and fair dealing. "If the employer merely disputes his liability under the contract by asserting in good faith and with probable cause that good cause existed for discharge, the implied covenant is not violated and the employer is not liable in tort. (Seaman's, supra, 36 Cal.3d at p. 770.) If, however, the existence of good cause for discharge is asserted by the employer without probable cause and in bad faith, that is, without a good faith belief that good cause for discharge in fact exists, the employer has tortiously attempted to deprive the employee of the benefits of the agreement, and an action for breach of the implied covenant of good faith and fair dealing will lie." (Id. at p. 1171; see also Khanna v. Microdata Corp., supra, 179 Cal. App. 3d 250, 262, 263 [breach of good faith and fair dealing established by proof of employer's bad faith action extraneous to the contract coupled with intent to frustrate the employee's contract rights].) Thus, under the Koehrer standard no bad faith action would lie unless the wrongful discharge was both willful and malicious.

As the majority candidly states, its concern is not so much with the workability of the foregoing standard as the fact that "any firing ... could provide the basis for a pleading alleging the discharge was in bad faith under the cited standards." (Maj. opn. at p. 699.) (Italics added.) In other words, such claims should be denied because otherwise courts would experience a "flood of litigation." The floodgate argument is no more plausible now, however, than it was in Klein v. Klein, supra, 58 Cal. 2d 692, where this court, in abolishing the rule of spousal immunity for negligent torts, stated: "'[C]ourts should not decline to entertain a meritorious action ... because of the dubious apprehension that in some future case [frivolous claims] may become the subject of litigation.'" (Id. at p. 694, quoting Spellens v. Spellens (1957) 49 Cal. 2d 210, 241 [317 P.2d 613] (conc. and dis. opn. of Schauer, J.).) Courts must deal with each case on its merits, "whether there be few suits or many; the existence of a multitude of claims merely shows society's pressing need for legal redress." (Dillon v. Legg, supra, 68 Cal.2d at p. 735, fn. 3.) That need has never been greater than in cases of malicious termination of employment. [47 Cal. 3d 723]

Conclusion

The majority's views of the bad faith issue reflect one overriding concern -- that of judicial overreaching. It is a familiar concern.

The power of the courts as both guardian and expositor of the common law is considerable. To be effective, such power must be exercised with restraint. Judge-made law must progress incrementally, "interstitially" as Justice Holmes said, filling the legislative gaps, neither too far ahead of nor too far behind the body politic.

Still, Justice Cardozo believed, as his landmark common law decisions attest, that the "fissures in the common law are wider than the fissures in a statute ...." (Cardozo, The Nature of the Judicial Process (1921) p. 71.) His point, it appears, is that within the framework of the traditional common law responsibilities of the judiciary, the courts may respond to changing social needs without deference to legislative action. As Professor Corbin said: "It is the function of our courts to keep the doctrines up to date with the mores by continual restatement and by giving them a continually new content. This is judicial legislation, and the judge legislates at his peril. Nevertheless, it is the necessity and duty of such legislation that gives to judicial office its highest honor; and no brave and honest judge shirks the duty or fears the peril." (Corbin, The Offer of an Act for a Promise (1920) 29 Yale L.J. 767, 771- 772, italics added.)

The majority's implicit fears of judicial activism are unfounded. We overstep no institutional bounds or constitutional constraints in recognizing that a willful and malicious termination of employment is so offensive to community values that it may give rise to tort remedies. On the contrary, such a holding would be entirely consistent with our judicial function, and with the great common law tradition of this court.

MOSK, J.

I dissent.

I am in agreement with the opinions of Justices Broussard and Kaufman with one significant exception: I am unwilling to accept their concurrence in part I of the majority opinion.

When an employee learns that one in a supervisorial position is an embezzler, he has the choice of two immediate courses of action. He can remain silent and thus avoid the enmity of the embezzler and embarrassment to the employer. That apparently is the approach preferred by my colleagues in order to assure the employee's retention of his job. Or, as a dutiful employee concerned with the image of his company, he can report [47 Cal. 3d 724] his knowledge to the employer. That is the course of action I would encourage.

My colleagues insist that reporting the presence of an embezzler to an employer is solely to the benefit of the employer. While undoubtedly it is to the employer's benefit, it is not exclusively so. It is my opinion that such action -- i.e., advising a state-created corporation of the employ in a supervisorial position of a person chargeable with a potential felony -- is in the best interests of society as a whole, and therefore covered by the public policy rule.

Under Labor Code section 1102.5, subdivision (b), an employer is prohibited from retaliating against an employee for disclosing information to a law enforcement agency when there is reasonable cause to believe a violation of state or federal laws has been committed. It seems incongruous to permit retaliation and discharge when the employee chooses to go directly to his employer with the information, rather than to circumvent the employer, go behind his back and directly to a public agency. In either event, it seems clear to me that the law and public policy are implicated.

With that one exception, I agree with the persuasive opinions of Justices Broussard and Kaufman.

FN 1. In September 1983, after plaintiff's discharge, Kuhne pleaded guilty in federal court to a felony count of embezzlement.

FN 2. Throughout its brief, defendant refers to a number of counterallegations. In particular, defendant alleges that plaintiff's true motive was an "attempt to oust his supervisor" and that plaintiff rejected an "opportunity to transfer laterally to a new position at company headquarters" before his discharge. Since this appeal arises on demurrer, however, we must assume the truth of the allegations in the complaint, and so these alleged facts are not properly before us.

FN 3. For a discussion of the origin of the employment-at-will doctrine, and its subsequent evolution through legislation and judicial decision, see generally Blades, Employment At Will vs. Individual Freedom: On Limiting the Abusive Exercise of Employer Power (1967) 67 Colum. L. Rev. 1404, 1416-1419; Blumrosen, Settlement of Disputes Concerning the Exercise of Employer Disciplinary Power; United States Report (1964) 18 Rutgers L. Rev. 428, 432-433; Comment, Employment-At-Will -- Employers May Not Discharge At-Will Employees For Reasons That Violate Public Policy (1986) Ariz. St. L.J. 161, 164-167; Note, Protecting At Will Employees Against Wrongful Discharge: The Duty to Terminate Only in Good Faith (1980) 93 Harv. L. Rev. 1816, 1824-1828 (hereafter Protecting At Will Employees).

FN 4. Even where employment is at will, numerous federal and state statutes already impose express limitations on the right of an employer to discharge at will. Legislation, ranging from the National Labor Relations Act (29 U.S.C. § 158(a)(1),(3),(4) [precluding discharge for union activity, protected concerted activity, filing charges and testifying under the act]) to title VII of the Civil Rights Act of 1964 (42 U.S.C. §§ 2000e-2, 2000e-3(a) [prohibiting discharge on the basis of race, color, religion, sex or national origin or for exercising rights under the act]), and state statutes precluding discharge for filing workers' compensation claims or in violation of state civil rights statutes, significantly circumscribe the at-will doctrine. (See Mauk, Wrongful Discharge: The Erosion of 100 Years of Employer Privilege (1985) 21 Idaho L. Rev. 201, 226-227, fns. 109-110 [hereafter Mauk].)

FN 5. A number of California decisions have followed Petermann to bar discharge of at-will employees in violation of state policies governing labor-management relations. (Montalvo v. Zamora (1970) 7 Cal. App. 3d 69 [86 Cal. Rptr. 401]; Wetherton v. Growers Farm Labor Assn. (1969) 275 Cal. App. 2d 168 [79 Cal. Rptr. 543]; Glenn v. Clearman's Golden Cock Inn, Inc. (1961) 192 Cal. App. 2d 793 [13 Cal. Rptr. 769].)

FN 6. We observed that courts in California and sister states had shown a willingness to grant tort damages in such instances. (See Tameny, 27 Cal.3d at p. 177; Kouff v. Bethlehem-Alameda Shipyard (1949) 90 Cal. App. 2d 322 [202 P.2d 1059].)

FN 7. A comparison of the manner in which contracts for illegal purposes are treated is useful. (See e.g., Civ. Code, §§ 1441, 1598, 1608, 1667.) If an employer and employee enter into a contract in which the employee agrees to perform an illegal act, and either party breaches, the courts will not enforce the contract. In the Tameny situation, however, typically the employer requires the employee to perform an act in contravention of public policy and then terminates the employee for failure to do so, or penalizes the employee for acting in accordance with law. When such a termination occurs, the nature of the employee's relationship with the employer, whether at will or contractual, is essentially irrelevant. What is vindicated through the cause of action is not the terms or promises arising out of the particular employment relationship involved, but rather the public interest in not permitting employers to impose as a condition of employment a requirement that an employee act in a manner contrary to fundamental public policy.

FN 8. We also quoted with approval Petermann's statement that "'[t]he public policy of this state as reflected in the Penal Code sections referred to above would be seriously impaired if it were to be held that one could be discharged by reason of his refusal to commit perjury.'" (Petermann, supra, 174 Cal.App.2d at p. 189, quoted in Tameny, supra, 27 Cal.3d at p. 173.)

FN 9. Defendant disputes this characterization of plaintiff's motives and conduct. In reviewing a judgment following the sustaining of a demurrer without leave to amend, however, we must accept the truth of the matters pleaded (see ante, pp. 662-663), and take into account the possibility of amendment consistent with that pleading. (Blank v. Kirwin (1985) 39 Cal. 3d 311, 318 [216 Cal. Rptr. 718, 703 P.2d 58].)

FN 10. By contrast, Labor Code section 1102.5, subdivision (b) prohibits an employer from retaliating "against an employee for disclosing information to a government or law enforcement agency, where the employee has reasonable cause to believe that the information discloses a violation of state or federal statute, or violation or noncompliance with a state or federal regulation."

FN 11. As noted, Tameny insisted that the public policy basis for the cause of action must be "firmly established" (27 Cal.3d at p. 172), "fundamental" (id., at p. 176), and "substantial" (id., at p. 177, quoting Harless v. First Nat. Bank in Fairmont (1978) 162 W.Va. 116 [246 S.E.2d 270]). Cases of other jurisdictions similarly require a "clearly mandated public policy." (See, e.g., Palmateer v. International Harvester Co. (1981) 85 Ill. 2d 124 [421 N.E.2d 876, 878]; Parnar v. Americana Hotels, Inc. (1982) 65 Hawaii 370 [652 P.2d 625, 630-631]; Thompson v. St. Regis Paper Co. (1984) 102 Wn.2d 219 [685 P.2d 1081, 1088-1089]; Adler v. American Standard Corp. (1981) 291 Md. 3 [432 A.2d 464, 472-473].)

FN 12. The absence of a distinctly "public" interest in this case is apparent when we consider that if an employer and employee were expressly to agree that the employee has no obligation to, and should not, inform the employer of any adverse information the employee learns about a fellow employee's background, nothing in the state's public policy would render such an agreement void. By contrast, in the previous cases asserting a discharge in violation of public policy, the public interest at stake was invariably one which could not properly be circumvented by agreement of the parties. For example, in Tameny, supra, 27 Cal. 3d 167, a contract provision purporting to obligate the employee to comply with an order of the employer directing the employee to violate the antitrust laws would clearly have been void as against public policy, and in Petermann, supra, 174 Cal. App. 2d 184, a contract provision which purported to obligate the employee to commit perjury at the employer's behest would just as obviously have been invalid. Because here the employer and employee could have agreed that the employee had no duty to disclose such information, it cannot be said that an employer, in discharging an employee on this basis, violates a fundamental duty imposed on all employers for the protection of the public interest.

FN 13. The California case most closely in point is Read v. City of Lynwood (1985) 173 Cal. App. 3d 437 [219 Cal. Rptr. 26]. Plaintiff, a city employee, protested the award of a city contract to a developer who, she had heard, had once attempted to bribe a fellow employee. That protest and other actions led the city to eliminate plaintiff's regular position and to terminate her employment. The Court of Appeal reversed a dismissal of her suit for further consideration of questions relating to the elimination of her position, but upheld dismissal of the Tameny cause of action.

We have discovered no case upholding a cause of action on facts comparable to the present case.

FN [14]. We observe there is nothing intrinsically unfair or uncommon about a contract that permits the employee to leave at will, but limits the employer's right of discharge. As explained in Pugh, supra, 116 Cal. App. 3d 311, 325, "A contract which limits the power of the employer with respect to the reasons for termination is no less enforcible because it places no equivalent limits upon the power of the employee to quit his employment. 'If the requirement of consideration is met, there is no additional requirement of ... equivalence in the values exchanged, or "mutuality of obligation."' (Rest.2d Contracts, § 81 (Tent. Draft No. 2, 1965); 1A Corbin on Contracts (1963) § 152, pp. 13-17; see Chinn v. China Nat. Aviation Corp. (1955) 138 Cal. App. 2d 98 [291 P.2d 91]; Toussaint v. Blue Cross & Blue Shield of Mich. (1980) 408 Mich. 579, 600 [292 N.W.2d 880].)" As we shall discuss further in the next portion of the opinion, there is no reason why standard contract principles should not apply in the employment context.

FN 15. Although the statute of frauds defense was not argued or even anticipated in Pugh, supra, 116 Cal. App. 3d 311, or Cleary, supra, 111 Cal. App. 3d 443, we note that the rule of Newfield, as applied to plaintiff, would indirectly circumvent those cases by holding that any claim based on the breach of an oral or implied promise to discharge only for cause is unenforceable unless the employee has been employed for less than one year. Because Cleary had 18 years service and Pugh 32, both their claims would have been defeated by an affirmative defense based on Newfield's view of the statute of frauds.

FN 16. The Court of Appeal in Gray v. Superior Court, supra, 181 Cal. App. 3d 813, 822, noted the conflict between Newfield and White Lighting and, pending our decision in the present appeal, declined to follow Newfield. Another Court of Appeal decision, Steward v. Mercy Hospital (1987) 188 Cal. App. 3d 1290, 1295 [233 Cal. Rptr. 881], declared that Newfield represented a minority view, and held that under the majority view enforcement of an oral employment contract was not barred by the statute of frauds if, within one year, the employee could terminate the contract or be discharged for cause.

We also note that the Ninth Circuit recently held in Robards v. Gaylord Bros., Inc. (9th Cir. 1988) 854 F.2d 1152, 1154-1155, that under California law the statute of frauds would not defeat an action for breach of an employment contract.

FN 17. Compare, for example, our decision in Phillippe v. Shappell Industries (1987) 43 Cal. 3d 1247 [241 Cal. Rptr. 22, 743 P.2d 1279]. There, we were called on to apply to Civil Code section 1624, subdivision (d), which requires that an agreement with an agent or broker, or other person to purchase, sell, or lease real estate for over a year, or to find a purchaser or seller or a lessee or lessor for a lease of more than one year be in writing. As we observed in Phillippe, "The Courts of Appeal have consistently held, with two narrow exceptions not present here, that a licensed broker may not assert estoppel against a statute of frauds defense in an action to recover a commission under an oral employment agreement." (43 Cal.3d at p. 1260.) The histories of subdivisions (a) and (d) of section 1624 have taken very different courses and the two subdivisions apply in different circumstances.

FN 18. A second source of conflict with White Lighting concerns the Court of Appeal's use of a retrospective analysis that infers the original intent of the parties from the ultimate duration of the employment at the time of the discharge. The applicability of the statute of frauds must be analyzed prospectively, based on the intentions of the parties and the terms of the agreement at the time it is made. "'To fall within the words of the provision, therefore, the agreement must be one of which it can truly be said at the very moment it is made, "This agreement is not to be performed within one year"; in general, the cases indicate that there must not be the slightest possibility that it can be fully performed within one year.'" (White Lighting, supra, 68 Cal.2d at p. 343, fn. 2, quoting 2 Corbin on Contracts (1950) § 444, at pp. 534-535 [italics added in White Lighting].) The Court of Appeal's retrospective approach is erroneous because it effectively changes the rule of White Lighting from a "possibility of performance" to a "probable expectations" test.

FN 19. Newfield, supra, 156 Cal. App. 3d 440, is disapproved to the extent that it is inconsistent with this opinion.

FN 20. Plaintiff alleges defendant maintained written "Guidelines for Termination" that required good cause for discharge of an employee, and that plaintiff understood these guidelines applied to him. If he had further alleged that the parties expressly agreed that these guidelines governed his employment, he could state a cause of action for breach of an express oral contract. He has made no such allegation.

FN 21. Comment (a) to Restatement Second of Contracts, section 4, explains, "[j]ust as assent may be manifested by words or other conduct, sometimes including silence, so intention to make a promise may be manifested in language or by implication from other circumstances. ..." Comment (a) to Restatement Second of Contracts, section 19, adds, "there is no distinction in the effect of the promise whether it is expressed in writing, or orally, or in acts, or partly in one of these ways and partly in others."

FN 22. After discussing Lord and similar cases, the Michigan Supreme Court observed that the issue in those actions "was whether, assuming a contract for 'permanent' employment, that employment was terminable at the will of the employer." (Toussaint, supra, 292 N.W.2d at p. 890.) In other words, those courts were considering not whether an employee could be discharged at will, but whether he could be discharged at all.

FN 23. A few recent cases, including Shapiro v. Wells Fargo Realty Advisors, supra, 152 Cal. App. 3d 467, can be interpreted to preclude enforcement of an implied-in-fact modification of an on-going employment agreement when some express written provision insists on the employee's at-will status. (See also Crain v. Burroughs Corp. (C.D.Cal. 1983) 560 F. Supp. 849 [applying Cal. law]; Crossen v. Foremost-McKesson, Inc. (N.D.Cal. 1982) 537 F. Supp. 1076 [same].) This, however, is not such a case. The only two writings revealed by plaintiff's complaint -- an agreement barring plaintiff's employment with a competitor, and one requiring his assignment of all computer-related information and innovations for one year after termination -- are silent on permissible bases for discharge and arguably are consistent with both a promise not to terminate except for good cause under defendant's termination guidelines and defendant's acceptance of independent consideration to support such a promise.

FN 24. We need not here resolve questions concerning the measure of damages in a wrongful discharge action based on breach of contract. We do not know what damages plaintiff is seeking to recover nor what particular evidence he may present in support of such damages, and therefore the issue is not before us.

FN 25. At times certain breaches of contract have been deemed economically desirable (see Putz, supra, 21 U.S.F. L.Rev. at pp. 430-432; Comment, Reconstructing Breach of the Implied Covenant of Good Faith and Fair Dealing as a Tort (1985) 73 Cal.L.Rev. 1291, fn. 3 [hereafter Reconstructing Breach]). As the reporter's notes to Restatement Second of Contracts, chapter 16, section 344 et seq., pages 101-102, state, "a breach of contract will result in a gain in 'economic efficiency' if the party contemplating breach evaluates his gains at a higher figure than the value that the other party puts on his losses, and this will be so if the party contemplating breach will gain enough from the breach to have a net benefit even though he compensates the other party for his resulting loss."

FN 26. In only three cases outside of California have courts held that a breach of the covenant of good faith and fair dealing gives rise to tort damages. In K Mart Corp. v. Ponsock (1987) 103 Nev. 39 [732 P.2d 1364], the Nevada Supreme Court followed Koehrer, supra, 181 Cal. App. 3d 1155, to hold that tort damages were available in an employment termination case. The court limited its holding to "this fact-specific instance of discharge by a large, nation-wide employer of an employee in bad faith for the improper motive of defeating contractual retirement benefits." (732 P.2d at p. 1370.) In Carter v. Catamore Co. (N.D.Ill. 1983) 571 F. Supp. 94, 97, an Illinois federal district court, applying Rhode Island law and citing a Rhode Island case involving an insurance contract, summarily and without analysis, stated that tort damages were available for breach of the covenant in an employment case. Carter has not been cited by any Rhode Island court.

Finally, in Dare v. Montana Petroleum Marketing Co. (1984) 212 Mont. 274 [687 P.2d 1015, 1020], the court apparently held that tort liability would lie for breach of an employment contract's implied covenant of good faith and fair dealing. Its formulation bears close scrutiny: "Whether a covenant of good faith and fair dealing is implied in a particular case depends upon objective manifestations by the employer giving rise to the employee's reasonable belief that he or she has job security and will be treated fairly." (Ibid.) As the concurring opinion points out, generally the covenant is deemed implicit in every employment contract. (Id., at p. 1021, Morrison, J., conc.) The Montana standard in fact looks very like one applicable to ascertaining the existence of an implied-in-fact contractual term not to discharge except for good cause. Moreover, subsequent to the Dare decision, the Montana Legislature enacted a new statutory scheme governing available remedies for discharge. (See post, at p. 695, fn. 31.)

Hall v. Farmers Ins. Exchange (Okla. 1986) 713 P.2d 1027, 1031, arguably also suggested that damages beyond contract remedies might be available. But, although at one point citing a broad rule for damages recoverable for breach of employment contracts, the court ultimately limited the measure of damages to "the predictable, quantifiable amount of future income which [plaintiff] was entitled to receive as renewal premiums." (Ibid.)

Other states have explicitly rejected the notion of either tort damages for breach of the implied covenant or application of the implied covenant itself in the context of employment relationships (see, e.g., Arco Alaska, Inc. v. Akers (Alaska 1988) 753 P.2d 1150, 1153-1154 [expressly rejecting tort recovery of punitive damages and limiting awards to contract damages in cases involving breach of the covenant]; Morriss v. Coleman Co., Inc. (1987) 241 Kan. 501 [738 P.2d 841, 849-851] [implied covenant should not apply to at-will employment contracts]; Hillesland v. Federal Land Bank Ass'n (N.D. 1987) 407 N.W.2d 206, 210-215 [same]; Wagenseller v. Scottsdale Memorial Hosp. (1985) 147 Ariz. 370 [710 P.2d 1025, 1040] [no tort action]; Hunt v. IBM Mid America Employees Federal (Minn. 1986) 384 N.W.2d 853, 858 [no implied covenant read into employment contracts]; Thompson v. St. Regis Paper Co., supra, 685 P.2d 1081, 1086 [same]; Martin v. Federal Life Ins. Co., supra, 440 N.E.2d 998, 1006 [no tort action]). Still others have impliedly denied tort damages by limiting recovery for breach of the implied covenant to benefits earned under the contract. (E.g., Cook v. Alexander and Alexander (1985) 40 Conn. Super. 246 [488 A.2d 1295, 1297], citing Magnan v. Anaconda Industries, Inc. (1984) 193 Conn. 558 [479 A.2d 781]; Maddaloni v. Western Mass. Bus Lines, Inc., supra, 438 N.E.2d 351, 356.)

Despite the concurring and dissenting justices' contentions to the contrary, almost every court considering the issue outside of California has either totally rejected applying the convenant of good faith and fair dealing to employment contracts, or has limited recovery for breaches of the covenant to contract damages.

FN 27. Contrary to Justice Broussard's suggestion (see opn. by Broussard, J., post, at p. 704, our statements in Seaman's were far from a definitive signal of approval for a tort remedy for breach of the covenant in employment cases. If anything, the reference highlighted the fact that this question remained to be decided by this court.

FN 28. Justice Broussard suggests that "unanimous agreement" among Courts of Appeal and some commentators on the existence of tort liability for a breach of the covenant of good faith and fair dealing has led to reliance by the public which we should be loath to disturb. (See opn. by Broussard, J., post at p. 706.) There has, however, clearly and indisputably, been no holding by this court that such a cause of action exists. If we were to follow the dissent's urging that we should therefore leave this area of law untouched, we would be abdicating our role. Trial courts are of course bound by the decisions of the Court of Appeal. But decisions of the lower appellate court are in no way binding upon this court which is free at any time to overrule lower court interpretations of questions of law and reach a different conclusion. (Cf. INS v. Chadha (1983) 462 U.S. 919, 944-945 [77 L. Ed. 2d 317, 340-341, 103 S. Ct. 2764] [unicameral legislative veto power over executive agency decisions violates constitutional separation of powers; decision affected almost 200 congressionally created veto-type procedures enacted over 50 years]; dis. opn. by White, J. at pp. 976-977 [77 L.Ed.2d at pp. 360-361] [arguing that differing opinions on constitutionality of legislative veto and Congress's reliance on its propriety for over 50 years weighed heavily in favor of approving the disputed procedure].) Engaging in interpretation of an area of law for the first time hardly amounts to a "radical restructuring" of the law as Justice Broussard's opinion expansively suggests.

FN 29. The Court of Appeal in Wallis v. Superior Court (1984) 160 Cal. App. 3d 1109 [207 Cal. Rptr. 123], specifically undertook an analysis of the features of an insurance contract which distinguished it from other ordinary commercial relationships. It concluded that the employment relationship was analogous to that of insurer and insured and justified the availability of tort relief because the parties "were in an inherently unequal bargaining position because of plaintiff's need for the money and his relative lack of skills outside the furniture manufacturing area. Second, plaintiff entered the contract to secure financial stability and peace of mind. Third, ordinary contract damages offered no incentive for defendant not to breach. ... Fourth, plaintiff was in an extremely vulnerable position because of his age, lack of other work skills and financial responsibilities, and fifth, defendant was aware of plaintiff's position." (Id., at p. 1119.)

FN 30. In the employment relationship, the employer is already barred by law from, and the employee can obtain relief for, discriminatory discharges based on age, sex, race, and religion. Similarly, the employee is protected from discrimination based on the exercise of rights under the workers' compensation laws or for engaging in union activities. He can gain relief if he is terminated in order for the employer to avoid payment of certain benefits. The employee may sue in tort for discharges based on breaches of public policy.

FN 31. Legislatures, in making such policy decisions, have the ability to gather empirical evidence, solicit the advice of experts, and hold hearings at which all interested parties may present evidence and express their views, and legislatures in three jurisdictions have enacted measures addressing terms of discharge. Puerto Rico provides employees serving under an indefinite term contract who are discharged without good cause: 1) any backpay due, 2) one month's salary, and 3) "an additional progressive indemnity equivalent to one full week for each year of service." (P.R. Laws Ann., tit. 29, § 185a (Supp. 1983).) In the Virgin Islands, the permissible grounds for discharge are described by a statute which also provides for reinstatement and backpay remedies under certain circumstances. In addition, an employee may bring an action for compensatory and punitive damages. (V.I. Code Ann., tit. 24, ch. 3, §§ 76-79 (1986).) Montana has a "Wrongful Discharge From Employment Act" which states employment with no term specified is at will. (Mont. Code Ann., §§ 39-2-901 -- 39-2-905; 39-2-911-39-2-914.) Nonetheless, a discharge is wrongful under three circumstances. First, if it is in retaliation for refusal to violate or to report a violation of public policy. Public policy is defined as a policy covering health, safety or welfare and established by constitutional provision, statute, or administrative rule. Second, the employee had successfully completed his probationary period and the discharge was not for good cause. Third, the discharge was in violation of express terms of a written personnel policy. Remedies are limited to wages and fringe benefits for no more than four years after date of discharge, and include interest. Punitive damages may be obtained only by a showing by clear and convincing evidence of actual fraud or malice by the employer. The remedies provided preempt any common law remedies.

Many European and other western countries have job protection laws, some providing separate tribunals or alternative methods of dispute resolution to handle claims which arise. (See, Note, Protecting At Will Employees, supra, 93 Harv. L. Rev. at pp. 1836-1837; Gould, Stemming the Tide, supra, 13 Emp.Rel.L.J. at pp. 408; Unfair Dismissal (1984) 20 Stan.J.Internat.L., issue 2.)

FN 32. "The principal goals of [legislation regulating the workplace in the last 75 years] have been first, to promote unionization as a countervailing force against employer power and control; second, to establish a minimum level of economic entitlement for workers; third, to combat discrimination against specific groups in hiring and dismissals; fourth, to protect employee health and safety; and fifth, to guarantee a minimum level of security for retirement and for the survivors of wage earners." (Note, Protecting At Will Employees, supra, 93 Harv. L. Rev. 1816, 1827, fns. omitted.)

FN 33. The generally predictable and circumscribed damages available for breaches of contract reflect the importance of this value in the commercial context. In the employment context specifically, Professor Gould observes, "[t]he cost of lawsuits that respond to a discharge, as measured by jury awards and settlements, has also increased geometrically and is beginning to draw concern from the business community. ... [¶] [T]he awards [in California wrongful termination suits] actually exceeded settlement demands by the employees' lawyers by 187 percent." (Gould, Stemming the Tide, supra, 13 Emp.Rel.L.J. at pp. 405-406, fns. omitted, italics in original; see also Note, Protecting At Will Employees, supra, 93 Harv. L. Rev. 1816, 1834, 1835 ["employers may fear that imposition of liability threatens the fundamental prerogatives of management: to control the workplace and to retain only the best-qualified employees"; "[a]nother concern is that the threat of increased liability will impair firms' ability to respond flexibly to changing economic conditions"].)

FN 34. See Burton, supra, 94 Harv. L. Rev. 369, 403 ("The good faith performance doctrine, like contract law generally, functions to support the market").

FN 35. One attempt to formulate a test for tortious breach of the implied covenant demonstrates the problem. Under its proposed standard, the elements of a tortious breach which a plaintiff must prove are: "(1) assertion of a right or denial of an obligation (2) made in bad faith (with actual knowledge that the claim or denial has no foundation) and unreasonably (where a reasonable person under the circumstances would find the claim or denial groundless) (3) that obstructs the injured party's ability to receive the substitutionary value of the agreement." (Comment, Reconstructing Breach, supra, 73 Cal.L.Rev. at p. 1305.) The commentator describes the two prongs of his bad faith test as subjective and objective, and acknowledges, "[s]ubjective bad faith is difficult to distinguish from mere doubt about the legitimacy of an act or claim." (Id., at pp. 1304-1305.) Unfortunately, even if the suggested test would have some restrictive value on the ability of plaintiffs to recover damages, it does not serve to limit initiation and prosecution of litigation based on almost any discharge.

Also interesting is the commentator's recognition that punitive damages are inappropriate. His argument for imposition of tort liability is that it "provides more adequate compensation for victims of bad faith conduct." (Id., at p. 1311.) As to punitive damages, however, he states, "they are not appropriate for breach of the implied covenant of good faith and fair dealing. Tort damages already remedy the inadequate compensation and deterrence of contract damages. ... Additional punitive damages are unnecessary to redress undercompensation for harm caused by a breach of the implied covenant." (Id., at pp. 1325-1326, fn. omitted.) Careful scrutiny thus reveals that the proposed approach is essentially a variant on proposals to expand contract damages in order to provide sufficient compensation.

FN 36. Moreover, as Professor Gould observes, "Juries in these cases often impose liability and large damage awards according to their own standards of fairness rather than the legal instructions provided by the judge. ... [¶] Jurors can easily identify with the worker who has received a pink slip." (Gould, Stemming the Tide, supra, 13 Emp.Rel.L.J. at pp. 406-407; see also Epstein, In Defense of the Contract At Will (1984) 51 U. Chi. L. Rev. 947, 970 [arguing juries are ill-suited to the task of determining an employer's motive or purpose for discharge].)

FN 37. Unlike collective bargaining agreements that contain "screening mechanisms" whereby "unions sift out grievances that are viewed as unmeritorious or less important" (Gould, Stemming the Tide, supra, 13 Emp.Rel.L.J. at p. 423), the proposed tort action essentially has no entry-level limitation (cf. Huber, supra, 841 F.2d at pp. 985-986).

FN 38. In a contract action asserting breach of a covenant not to discharge except for good cause, the plaintiff must demonstrate an agreement not to terminate except for good cause and that the employer lacked good cause for the discharge. The Khanna and Koehrer formulations add the element of "bad faith" but fail to explain how courts are to distinguish the "bad faith" giving rise to tort recovery, from the lack of "good faith" giving rise to contract damages.

FN 39. Moreover, with regard to an at-will employment relationship, breach of the implied covenant cannot logically be based on a claim that a discharge was made without good cause. If such an interpretation applied, then all at-will contracts would be transmuted into contracts requiring good cause for termination, and Labor Code section 2922 would be eviscerated. This is not to say that the Legislature could not impose such a requirement in every employment contract. It has not done so, however, and the implied covenant should not be read as uniformly imposing such a contractual term. Thus, in Wagenseller v. Scottsdale Memorial Hosp., supra, 710 P.2d at page 1040, the court properly stressed, "[w]hat cannot be said is that one of the agreed benefits to the at-will employee is a guarantee of continued employment or tenure." Because the implied covenant protects only the parties' right to receive the benefit of their agreement, and, in an at-will relationship there is no agreement to terminate only for good cause, the implied covenant standing alone cannot be read to impose such a duty. (See ibid.)

FN 40. The court stated, "[w]hile the specific nature of the obligations imposed by the implied covenant ... are dependent upon the nature and purpose of the underlying contract and the legitimate expectations of the parties arising from the contract [citations], those obligations are not the obligations that were consensually undertaken in the contractual provisions, and care must be taken in each case to determine whether the alleged breach is of an obligation imposed by law and thus a tort [citation] or breach of an obligation consensually created by the parties in the terms of the contract and thus simply a breach of contract." (181 Cal.App.3d at p. 1169.)

FN 41. As the Arizona Supreme Court recognized, "the implied-in-law covenant of good faith and fair dealing protects the right of the parties to an agreement to receive the benefits of the agreement that they have entered into. The denial of a party's rights to those benefits, whatever they are, will breach the duty of good faith implicit in the contract. Thus, the relevant inquiry always will focus on the contract itself, to determine what the parties did agree to." (Wagenseller v. Scottsdale Memorial Hosp., supra, 710 P.2d at p. 1040, italics added.)

FN 42. Clearly, supra, 111 Cal. App. 3d 443, and its progeny accordingly are disapproved to the extent that they permit a cause of action seeking tort remedies for breach of the implied covenant.

FN 43. We do not reach the issue of the retroactive or prospective application of our opinion. The parties have not briefed or argued the question and we will deal with the matter in a later case when we have the benefit of the views of counsel.

FN 1. Application of these precedents will not solve all the problems the majority create. There will be cases in which the probability of emotional distress is not present at the time of contracting, but becomes apparent only later, after the employee has put in years on the job and come to rely on it for his economic security. Most of the cases cited seem to limit recovery to damages foreseeable at the time of the contract. (But see Civ. Code, § 3300 [the measure of damages for breach of contract "is the amount which will compensate the party aggrieved for all the detriment proximately caused thereby, or which, in the ordinary course of things, would be likely to result therefrom"]; Overstreet v. Merritt (1921) 186 Cal. 494, 505 [200 P. 11] ["[o]ne who in bad faith violates his contract is liable for all damages traceable to the breach, including even those which could not be foreseen at the time of making the contract"].) Secondly, these precedents provide no basis for punitive damages, no matter how outrageous the employer's conduct, or how essential such damages may be to deter future wrongful conduct.

FN 2. Chief Justice Bird, in her concurring and dissenting opinion, wrote that tort remedies should have a broader scope in insurance and employment cases. She observed that "breach of an employment contract by the employer can, in some situations, cause severe harm to an employee's reputation and ability to find new employment. The harm caused cannot be undone by an award of backpay. Thus, employees may be entitled to expect that their contracts will not be breached for frivolous or improper reasons." (Seaman's, supra, 36 Cal.3d at p. 780.)

FN 3. The majority do not decide whether their decision abolishing the tort cause of action for bad faith discharge is retroactive. A pattern of reliance similar to that present here led this court to make its decision in Moradi-Shalal v. Fireman's Fund Ins. Companies (1988) 46 Cal. 3d 287, 305 [250 Cal. Rptr. 116, 758 P.2d 58], prospective.

FN 4. This court has granted review in 14 cases pending the decision of the present case. Most of those cases involve, among other issues, the question of a tort remedy for bad faith discharge, and in all such cases the Court of Appeal unanimously upheld the existence of a tort remedy.

FN 5. Consider, for example, the case of K Mart Corp. v. Ponsock (1987) 103 Nev. 39 [732 P.2d 1364]. Ponsock was a forklift driver with 10 years longevity, whose pension would vest in another 6 months. He discovered that the battery cover on his forklift needed painting. Finding a damaged (and thus unsaleable) can of spray paint, he used it to paint the forklift. Although other forklifts had been painted in a similar manner without any action against the employee in question, Ponsock was fired for "defacing" company property and "stealing" a can of paint. When he attempted to explain his conduct, he was excluded from the premises. After a long period of unemployment, he finally obtained a job as a laborer at half his previous wage with no benefits. This income was inadequate to meet mortgage payments, and Ponsock was forced to sell his home at a loss.

Traditional contract damages would not compensate Ponsock for the loss incurred on the sale of a home, or for emotional suffering. In addition, such damages would do nothing to deter further arbitrary actions. The employer would be required to compensate Ponsock for the difference between his prior wage and his present wage, but since Ponsock could be replaced by a lower wage worker whose pension would not vest for many years, the employer might profit from his wrong.

FN 6. A second significant similarity is that both insurance contracts and employment contracts arise from a context of disparity of bargaining power. Numerous cases have noted this disparity in insurance cases; it has led to the adoption of a general rule that insurance contracts are construed against the insurer. There are fewer cases in the employment context, but here the principle is embodied in a statutory finding that "the individual unorganized worker is helpless to exercise actual liberty of contract and to protect his freedom of labor, and thereby to obtain acceptable terms and conditions of employment." (Lab. Code, § 923.)

FN 7. All contracts, including contracts for employment at will, include a covenant of good faith and fair dealing, but the arbitrary discharge of an employee at will is not a breach of contract. (See discussion in maj. opn., ante, at p. 698, fn. 39.)

FN 8. The importance of predictability of damages, as the majority note (see ante, p. 683, fn. 25), is that it facilitates economically efficient breach. A party who can calculate damages can determine whether he can profit by breaching his contract, accepting liability in return for the benefits of breach. This attitude may be appropriate in a commercial context. It should not be condoned in the employer-employee relationship, where breach may cause injury beyond that of mere loss of income, injury which cannot easily be mitigated. It is difficult to summon sympathy for the employer who needs predictability of damages so he can calculate whether he will profit by firing his employee in breach of the employment contract.

FN 9. Of course, tort law is rife with subjective elements -- "malice," "willful misconduct," "reckless disregard," and, in insurance cases, "bad faith."

FN 10. Some cases and writers have suggested that the courts should also consider whether the employer acted reasonably. Since the majority abolish the entire cause of action, it is pointless now to decide that question. I would note only that the concept of reasonableness, like that of bad faith, is one familiar to tort law, and not generally considered so unpredictable or subjective as to justify denial of relief for injuries suffered.

FN 11. The majority's fear that plaintiffs will plead bad faith when they have no basis for that pleading seems exaggerated. I see no more reason to fear false pleading in this context than in any other cause of action. In any case, the majority's decision will not limit initiation and prosecution of litigation since plaintiffs will retain a contractual remedy. At best, it prevents plaintiffs from including false and groundless tort claims in their complaint, but does so only by also preventing them from including true and meritorious claims.

FN 12. Many of the commentators the majority cite to support their assertion that a tort remedy for bad faith discharge could have economic drawbacks are inapposite. A number of the articles deal solely with commercial contracts, and do not discuss a tort remedy for employees. (See Louderback & Jurika, Standards for Limiting the Tort of Bad Faith Breach of Contract (1982) 16 U.S.F. L.Rev. 187; Putz & Klippen, Commercial Bad Faith: Attorney Fees -- Not Tort Liability -- is the Remedy for Stonewalling, supra, 21 U.S.F. L.Rev. 419; Traynor, op. cit. supra, 8 Bus. L. News 1; Comment, Reconstructing Breach of the Implied Covenant of Good Faith and Fair Dealing as a Tort, supra, 73 Cal.L.Rev. 1291.) Others are primarily concerned with the rights of at-will employees, but the tort remedy for bad faith breach applies almost exclusively to cases involving employees with contractual protection against arbitrary discharge. (E.g., Gould, The Idea of the Job as Property in Contemporary America: The Legal and Collective Bargaining Framework, supra, B.Y.U. L. Rev. 885, 905.)

Most important, most of the commentators recognize the problem of inadequacy of contract damages and, if they do not support a tort remedy, support some alternative. I doubt that any would support the majority's act of abolishing tort damages without providing an alternative.

FN 13. The majority are not entirely consistent on this point, for they appear to leave open the possibility that the courts, without legislative action, could expand the measure of damages recoverable for discharge in breach of contract. I find it difficult to see why a common law court has greater ability to change the measure of damages in contract than to create a cause of action in tort.

But the practical problem with the majority's suggestion is that while they abolish the cause of action in tort, they only suggest the possibility of expanded contract damages. This is to trade a bird in the hand for the hope there will be one in the bush next year. Employees lose an adequate remedy, and may or may not get a replacement sometime.

FN 14. The majority disapprove Cleary v. American Airlines, Inc., supra, 111 Cal. App. 3d 443, "and its progeny." (Ante, p. 700, fn. 42.) The progeny of Cleary include Koehrer v. Superior Court, supra, 181 Cal. App. 3d 1155; Gray v. Superior Court, supra, 181 Cal. App. 3d 813; Khanna v. Microdata Corp., supra, 170 Cal. App. 3d 250; Rulon-Miller v. International Business Machines Corp., supra, 162 Cal. App. 3d 241; Shapiro v. Wells Fargo Realty Advisors, supra, 152 Cal. App. 3d 467; and Crosier v. United Parcel Service, Inc., supra, 150 Cal. App. 3d 1132.

FN 15. The majority refer to decisions of other jurisdictions, but present a misleading count. Five decisions of other jurisdictions uphold a tort cause of action (Huber v. Standard Ins. Co., supra, 841 F.2d 980; Carter v. Catamore Co., Inc. (N.D.Ill. 1983) 571 F. Supp. 94, 97 [R.I. law]; Dare v. Montana Petroleum Marketing Co. (Mont. 1984) 687 P.2d 1015, 1020; Gates v. Life of Montana Ins. Co. (1983) 205 Mont. 304 [668 P.2d 213]; K Mart Corp. v. Ponsock, supra, 732 P.2d 1364, 1369-1370); three reject that cause of action (Arco Alaska, Inc. v. Akers (Alaska 1988) 753 P.2d 1150, 1153-1154; Martin v. Federal Life Ins. Co. (1982) 109 Ill.App.3d 596 [440 N.E.2d 998, 1006]; Murphy v. American Home Products Corp. (1983) 58 N.Y.2d 293 [461 N.Y.S.2d 232, 238, 448 N.E.2d 86].) If we discount Huber as based on California law, the count stands at four to three. The majority compile a more extensive list of decisions by including cases involving quite different issues. A number of those decisions hold (contrary to California law) that there is no implied covenant of good faith in contracts for employment at will -- a holding which says nothing about whether tort damages would be awarded for the breach, in bad faith, of a contract barring termination without good cause.

If we count all cases, including California decisions and those based on California law, the current count is 13 cases upholding a tort cause of action and 3 rejecting it. This figure does not include the 10 or so additional cases upholding the cause of action depublished by the grant of review by this court and held pending the present decision.

FN 16. For the proposed legislation relating to wrongful discharge, see ante, page 714; for proposed legislation concerning insurance bad faith, see Moradi-Shalal v. Fireman's Fund Ins. Companies, supra, 46 Cal. 3d 287, 300 and pages 295-296 (Mosk, J., dis.).

FN 1. As the commentators have noted, the very term "implied covenant" is misleading, since it "evokes the notion of contract, and therefore the term 'duty' might be more appropriate for treating the violation of good faith and fair dealing as a tort." (Note, Reconstructing Breach of the Implied Covenant of Good Faith and Fair Dealing as a Tort (1985) 73 Cal.L.Rev. 1291, 1307, fn. 76.) The Restatement Second of Contracts, section 205, has adopted the language of duty: "Every contract imposes upon each party a duty of good faith and fair dealing in its performance and enforcement."

FN 2. In rejecting the doctrine of sovereign inmunity, we relied in part on earlier decisions in which we had abolished the doctrine of tort immunity of charitable institutions (Malloy v. Fong (1951) 37 Cal. 2d 356 [232 P.2d 241]; Silva v. Providence Hospital of Oakland (1939) 14 Cal. 2d 762 [97 P.2d 798]), "an immunity that was also claimed to be so firmly imbedded that only the Legislature could change it." (Muskopf v. Corning Hospital Dist., supra, 55 Cal.2d at p. 219.)