I dissent from the majority’s holding giving retroactive effect to the portion of their opinion in Foley v. Interactive Data Corp. *994(1988) 47 Cal.3d 654 [254 Cal.Rptr. 211, 765 P.2d 373] which abolished an employee’s tort cause of action for bad faith discharge.
In Moradi-Shalal v. Fireman’s Fund Ins. Companies (1988) 46 Cal.3d 287 [250 Cal.Rptr. 116, 758 P.2d 58] (hereafter Moradi-Shalal), this court overruled Royal Globe Ins. Co. v. Superior Court (1979) 23 Cal.3d 880 [153 Cal.Rptr. 842, 592 P.2d 329], the decision which established a cause of action under Insurance Code section 790.03 for bad faith delay in settling claims. We then turned to the question “whether our decision should apply to Royal Globe actions that already have been filed or litigated.” (46 Cal.3d at p. 305.) We answered that question in a single sentence: “Without implying any broad exception to the general rule of retrospectivity . . . and in the interest of fairness to the substantial number of plaintiffs who have already initiated their suits in reliance on Royal Globe, we hold that our decision overruling that case will not apply to those cases seeking relief under section 790.03 filed before our decision here becomes final.” (Ibid.) We should answer the question of the retroactivity of Foley v. Interactive Data Corp., supra, 47 Cal.3d 654 (hereafter Foley) in the same way, and for the same reason—concern for fairness to the substantial number of plaintiffs who have already initiated their suits in reliance on prior decisions recognizing a cause of action for bad faith discharge. The majority’s extended and convoluted analysis is entirely unnecessary.
Amazingly, the majority make no attempt to distinguish Moradi-Shalal, the most recent and pertinent precedent.1 They merely say that it was an exceptional case, as if only cases which follow general rules are based on legal reasoning, and those which craft exceptions to the rules are mere judicial caprice. Apparently the majority do not view Moradi-Shalal as a decision which established or applied a principled exception to the general rule of retroactivity, but instead see it as an aberrant, arbitrary decision without precedential value.
The majority accuse us of trying to transform Moradi-Shalal into a standard applicable to all cases.2 (Ante, p. 983.) But we do not claim Moradi-*995Shalal created a new rule of prospectivity. We think it applied an established standard, recognized by both majority and dissent, which prescribes retroactivity in the ordinary case, but permits prospectivity (or limited retroactivity) in the exceptional case in which a decision overturns established law upon which many have relied.3 We found that exception applicable in Moradi-Shalal. If the present case is indistinguishable from Moradi-Shalal, evenhanded justice requires that we find the exception applicable here.
But even if we follow the majority’s lead and disregard Moradi-Shalal, the standards established by prior California decisions and followed by the majority today compel the conclusion that the Foley holding should not be applied retroactively.4 Under that standard, we must first determine whether the decision in question creates a “new rule of law” (Donaldson v. Superior Court (1983) 35 Cal.3d 24, 36 [196 Cal.Rptr. 704, 672 P.2d 110]); if it does, we inquire into “(a) the purpose to be served by the new standards, (b) the extent of the reliance ... on the old standards, and (c) the effect on the administration of justice of a retroactive application of the new standards.” (Peterson v. Superior Court, supra, 31 Cal.3d 147, 152, quoting Stovall v. Denno (1967) 388 U.S. 293, 297 [18 L.Ed.2d 1199, 1203, 87 S.Ct. 1967].)
Did Foley create a new rule of law? The majority equivocate, but the answer is clear. The decision in Foley abolishing a cause of action for bad faith discharge established a new rule of tort law. That ruling was unsupported by California precedent. It overruled seven Court of Appeal opinions. (Cleary v. American Airlines, Inc. (1980) 111 Cal.App.3d 443 [168 Cal.Rptr. 722]; Crosier v. United Parcel Service, Inc. (1983) 150 Cal.App.3d 1132 [198 Cal.Rptr. 361]; Shapiro v. Wells Fargo Realty Advisors (1984) 152 Cal.App.3d 467 [199 Cal.Rptr. 613]; Rulon-Miller v. International Business Machines Corp. (1984) 162 Cal.App.3d 241 [208 Cal.Rptr. 524]; Khanna v. Microdata Corp. (1985) 170 Cal.App.3d 250 [215 Cal.Rptr. 860]; Gray v. Superior Court (1986) 181 Cal.App.3d 813 [226 Cal.Rptr. 570]; Koehrer v. Superior Court (1986) 181 Cal.App.3d 1155 [226 Cal.Rptr. 820].) It rejected the characterization of California law by the Ninth Circuit (Huber v. Standard Ins. Co. (9th Cir. 1988) 841 F.2d 980) and by numerous commentators *996(see Foley, supra, 47 Cal.3d 654, 706, and authorities there cited (Broussard, J., dis.)). It also impliedly repudiated the standard jury instructions used in wrongful termination cases. (See Cal. Jury Instns., Civ. (1988 Supp.) No. 10.54.)
The majority stress that Foley did not overrule a decision of this court, yet they recognize that questions of retroactivity arise not only when we overrule our own prior decision, but also in two other contexts: when the decision “disapproves a practice impliedly sanctioned by prior decisions of this court” or “disapproves a longstanding and widespread practice expressly approved by a near-unanimous body of lower-court authorities.”5 (People v. Guerra (1984) 37 Cal.3d 385, 401 [208 Cal.Rptr. 162, 690 P.2d 635]; see maj. opn. at pp. 980-981.) Both fit the circumstances of the present case.
The majority assert that the practice of permitting tort recovery for bad faith discharge was not “impliedly sanctioned” by this court because we never expressly approved the practice. But “impliedly sanctioned” does not mean “expressly approved”; it means “tacitly permitted.” When this court declines to review a consistent line of Court of Appeal decisions (finally granting review only when a decision adopts an extremely restrictive view of tort liability), permits those decisions to remain published, cites those cases without hint of disapproval, and goes so far as to approve the theory on which they rest,6 I conclude that we have impliedly sanctioned the practice in question.
The majority recognize that Foley overruled a consistent line of Court of Appeal cases, but maintain that the tort recovery permitted by those cases was not a long-standing practice.7 Nine years elapsed between the filing of Cleary v. American Airlines, Inc., supra, 111 Cal.App.3d 443, the first decision upholding a tort action for bad faith discharge, and the filing of our *997decision in Foley8 (Coincidentally, there were also nine years between Royal Globe Ins. Co. v. Superior Court, supra, 23 Cal.3d 880, and Moradi-Shalal.)9
How long is long enough? I suggest the answer, based on the interests served by a decision limiting retroactivity, is long enough that people can and do reasonably rely on the practice. The majority agree: “what is most relevant is not the mere fact of the ‘novelty’ of the rule adopted, but rather the reliance of parties on the preexisting state of the law.” (Ante, p. 986.) The preexisting rule permitting a tort cause of action for bad faith discharge related to a type of dispute, the assertedly wrongful firing of a worker, which arises frequently. It impacted upon every aspect of the dispute, from the employee’s decision whether to initiate an action (and the attorney’s decision whether to take the case) through, pleading, discovery, settlement negotiation, trial, and appeal. Under these circumstances, nine years, or even six, is sufficient to engender significant reliance.
Since the decision in Foley disapproved a practice supported by a longstanding line of lower court authority and impliedly sanctioned by this court, it represents the kind of dramatic break from prior practice which raises a serious question whether “fairness and public policy” preclude retroactivity. (Peterson v. Superior Court, supra, 31 Cal.3d 147, 152.) As we noted earlier (ante, p. 995), our inquiry into that question requires us to consider (1) the purpose to be served by the new rule, (2) the extent of reliance on the former rule, and (3) the effect on the administration of justice. (Peterson v. Superior Court, supra, 31 Cal.3d at p. 152; Casas v. Thompson (1986) 42 Cal.3d 131, 140 [228 Cal.Rptr. 33, 720 P.2d 921].)
We begin, as do the majority, with the purpose of the new rule. The majority identify two purposes served by the decision in Foley. The first is *998that of “giving effect to the true expectations of the parties to a contract.”10 (Ante, at p. 986.) This purpose plainly calls for a prospective application of Foley. Parties contract with reference to the law as it exists at the time of the contract. Any employer contracting to hire or retain a worker between 1980, when Cleary v. American Airlines, Inc., supra, 111 Cal.App.3d 443, was filed, and 1989, when Foley was filed, would presumably expect that he would be subject to tort damages if he fired the employee in bad faith. He could not reasonably assume, as the majority suggest, that his liability would be limited to the “benefit of the bargain.” The retroactive application of Foley does not give effect to the parties’ expectations; it gives the employer an immunity from tort damages he did not expect, and takes from the worker a remedy he thought he possessed.
Another purpose of the Foley decision, the majority say, was to enhance the “predictability of the consequences of actions related to employment contracts.” (Ante at p. 989, quoting Foley, supra, 47 Cal.3d at p. 696.) This purpose even more clearly points to a prospective application of the decision. One can only predict the future. Nothing we do can alter or facilitate past predictions. All we can do is fulfill or disappoint them. And the retroactive application of Foley will necessarily disappoint past predictions in which parties, relying on existing law, decided whether to file or defend suits, whether to accept or reject settlements, and made other decisions incident to litigation.
Under the format established by prior cases, a format we and the majority both follow, the next consideration in our inquiry whether considerations of fairness and public policy preclude retroactivity of the new rule is the extent of the reliance on the former rule. In my dissent in Foley, I explained the extensive reliance on the now-overruled Court of Appeal decisions: “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 exits. Many others [are] pending in the Court of Appeal.” (Foley, supra, 47 Cal.3d at p. 706 (Broussard, J., dis.).)
The majority did not dispute these facts in Foley and do not dispute them now. In fact the majority never discuss the extent of reliance on prior law. *999Yet despite recognizing that the extent of reliance is central to their opinion, the majority retreat to the periphery. They circle the issue warily, pointing out facts which might suggest that the reliance was a little less extensive than it might otherwise have been, or that persons who reasonably relied on preexisting law may have done so mindful that it is always possible that someday the law may be different. For their opinion to make logical sense, however, such tangential observations are not enough. The majority must claim either that there was no extensive reliance on pre-Foley cases, or that the reliance was unreasonable. They are too honest to make either claim, for they know that reasonably competent judges and attorneys relied extensively upon the then-controlling Court of Appeal decisions.
Among their peripheral observations is the suggestion that Foley did not deprive the employee of all remedies whatsoever.11 They do not discuss whether the remedies that remain will provide a worker discharged in bad faith with adequate recovery, or even with sufficient damages to enable him to hire counsel. Yet, when this issue was discussed in Foley the majority recognized that the remaining remedies might be insufficient. (See Foley, supra, 47 Cal.3d 654, 669-700.)12
The last consideration in our inquiry is the effect of retroactivity on the administration of justice. Here there can be no dispute that retroactive application of Foley will require retrial of many cases now pending on appeal. The only question is how many. The majority hope that in many cases the jury will have rendered separate verdicts on tort or contract damages, and suggest that in such cases the court could simply strike the tort verdict. Such an action, however, could be quite unfair to plaintiffs who, confident of their tort cause of action, made no attempt to seek jury instructions on contract damages for foreseeable emotional distress.13 *1000The majority’s surprising conclusion, that administration-of-justice considerations favor retroactive application of Foley, rests on its argument “that the Courts of Appeal never reached a consensus on what sufficed for a claim of tortious breach of the implied covenant in the employment context.”14 (Ante, at p. 992.) The Court of Appeal decisions, of course, did reach a consensus that an employee could sue in tort for bad faith discharge; they disagreed primarily on whether certain matters, such as longevity of service and an express policy not to terminate arbitrarily, were elements to the tort or merely evidence of bad faith. (See discussion in Huber v. Standard Ins. Co., supra, 841 F.2d 980, 984.) But when Court of Appeal opinions conflict, it is our function to resolve that conflict. In Moradi-Shalal, after abolishing the cause of action for bad faith refusal to settle, we went on to establish the essential requisites of that tort for those actions still viable under our prospective decision. We could have done the same thing in Foley. We could do it today. We cannot justify retroactive abolition of the cause of action for bad faith discharge by our own failure to resolve conflicts in the courts below and to settle the essential elements of that cause of action.
In sum, the majority opinion is riddled with problems, but two stand out. First, it offers no grounds for distinguishing Moradi-Shalal, which just a few months ago held that fairness to plaintiffs who had relied on prior law justified making our holding prospective. Second, all of the majority’s reasoning about reliance is as insubstantial as mist against the solid fact of extensive and reasonable reliance on pre-Foley precedent, a fact they cannot deny. The majority’s refusal to protect such reliance by making its decision in Foley prospective will be catastrophic to the wrongfully discharged worker who relied upon the unanimous line of pre-Foley decisions in seeking redress for his wrong.
Mosk, J., and Kaufman, J., concurred.
The majority note that Moradi-Shalal, 46 Cal.3d at page 305, quoted Peterson v. Superior Court (1982) 31 Cal.3d 147, 152 [181 Cal.Rptr. 784, 642 P.2d 1305], where we said that when a “statute has received a given construction by a court of last resort, and contracts have been made or property rights acquired in accordance with the prior decision, neither will the contracts be invalidated nor will vested rights be impaired by applying the new rule retroactively.” But Moradi-Shalal did not go on to claim that contracts were made or property rights acquired in reliance upon the prior construction of Insurance Code section 790.03, the statute at issue in that case. Instead, it referred to the “substantial number of plaintiffs who have already initiated suit” (46 Cal.3d at p. 305) in reliance upon the prior interpretation of the statute. The present case also involves reliance by plaintiffs who have initiated suits based on the prior law. I fail to see any distinction between reliance upon past precedent interpreting a statute and reliance upon past precedent establishing a common law rule.
The majority’s fear that our view would make prospectivity the rule and retroactivity the exception arises from their failure to realize that we would take into account not only the mere fact of reliance, but also the nature and extent of that reliance.
This standard has been applied without regard to the area of law at issue. It is immaterial that the present case involves retroactivity of a rule of tort law.
The majority review federal cases which reject this method of analysis and require that virtually all decisions changing the law of criminal procedure be given retroactive effect. (Ante, pp. 980-981.) These cases rest on a very different view of retroactivity than do controlling California decisions; they emphasize equal treatment of all litigants, past and present, and give little weight to such considerations as reasonable reliance of past litigants or the burden on the administration of justice. The majority decision pays tribute to the federal cases, but follows the mode of analysis used in California cases.
People v. Bustamante (1981) 30 Cal.3d 88, 102 [177 Cal.Rptr. 576, 634 P.2d 927], held that a rule permitting counsel to be present at preindictment lineups should be given prospective application, in spite of the fact that this court had expressly refused to decide the issue, because prosecutors, police and courts had relied on prior Court of Appeal decisions.
I refer to the footnote in Seaman’s Direct Buying Service, Inc. v. Standard Oil Co. (1984) 36 Cal.3d 752, 769, footnote 6 [206 Cal.Rptr. 354, 686 P.2d 1158], where we said that the employment relationship shares some of the same characteristics as the relationship between insurer and insured.
The majority also note that the Court of Appeal decisions did not agree on the elements of the tort, so that litigants were aware that the scope of the cause of action had not been definitively established. That would be a good argument why a decision clarifying the scope of the cause of action should be given retroactive effect. It has no relevance to a decision abolishing the entire cause of action, including those matters on which the Court of Appeal decisions unanimously agreed.
The majority speak of six years between Cleary and the date when we granted review in Foley on the theory that our decision to grant review was clear notice of the possibility that we might reject a tort cause of action. That date is premature; the Court of Appeal decision in Foley erroneously held plaintiff’s contract cause of action barred by the statute of frauds, and adopted the most restrictive view of tort recovery of any Court of Appeal decision. Our grant of review suggested, if anything, that we preferred a broader basis for tort recovery. A more realistic date for the foreseeability of our rejection of a tort remedy might be the reargument of Foley in April of 1987. But neither date is really significant, for until we decided Foley the Court of Appeal decisions, later overruled by Foley, were binding authority (see Auto Equity Sales, Inc. v. Superior Court (1962) 57 Cal.2d 450 [20 Cal.Rptr. 321, 369 P.2d 937]) on trial courts throughout the state.
The majority note that nobody argues that either the portion of Foley which discussed a tort cause of action for discharge in violation of public policy, or that portion which overruled Newfield v. Insurance Co. of the West (1984) 156 Cal.App.3d 440 [203 Cal.Rptr. 9], should be denied full retroactive effect. The discussion of discharge in violation of public policy did not overturn any prior rule or establish any new rule. Newfield was promptly rejected by other districts of the Court of Appeal; the period in which it could be spoken of as established law would be measured in months. By the time we decided Foley our decision to disapprove Newfield conformed to the majority view below.
I question the majority’s statement of the purpose of Foley; the parties’ expectation that liability for breach of contract will be measured by principles of contract law can be secured without changing rules of tort law. But the majority opinion regrettably fails to draw a clear distinction between a tort cause of action for bad faith discharge and a contractual cause of action for breach of the covenant of good faith and fair dealing.
In a somewhat misleading paragraph, the majority assert that “Foley did not change the nature of an employer’s obligation under the contract, including adherence to the implied covenant of good faith and fair dealing. The employer remains bound by all express and implied terms of the contract, and any breach thereof is still actionable. Foley simply changed the nature of the remedy available for the breach.” (Ante, p. 991.)
One reading this paragraph would think that the issue on which the court divided in Foley was one of alternative contract remedies. But in fact the justices agreed unanimously on the contractual duties of the parties and the remedies for breach. The issue which divided us was tort duties arising from the employer-employee relationship, and remedies for their breach. The majority opinion in Foley abolished both the duties and the remedies.
The majority contrast Foley to Moradi-Shalal, which they assert completely deprived the claimant of his cause of action for bad faith refusal to settle. But the claimant under Moradi-Shalal retained his original judgment or settlement from the tortfeasor, and lost only an action for additional damages against the insurer. This seems quite analogous to the discharged employee who retains a claim for breach of contract, but loses his tort remedy.
The question whether contract damages could include emotional distress was left undecided by the Foley majority (see 47 Cal.3d 654, 682, fn. 24); it is discussed in my dissent at pages 701-702.
Again I must dispute the majority’s characterization of the cause of action at issue; it is not a cause of action for breach of a contractual covenant, but for the discharge of an employee without a good faith belief in the right to do so.