Opinion

JOHNSON, J.

This is a case of first impression in California.

*1537 The principal issue before us is whether evidence of tax benefits which a plaintiff may have received because of the expense of a contract should be considered in determining compensatory damages for the defendant’s breach of that contract. For the reasons stated below, we hold that such evidence is irrelevant. 1

Facts and Proceedings Below

In 1980 the parties entered into contracts by which the respondent physician purchased and licensed computer hardware and software from the appellant for $50,009.40. Appellant, a computer hardware and software supplier, promised the equipment could be used for accounting and billing purposes at both the respondent’s medical practice and dialysis facilities. The equipment was installed; however, it did not function as promised.

In July 1982 the respondent filed causes of action for breach of contract, failure of consideration, restitution, breach of warranty, and unlawful business practices.

The parties agreed to a summary trial: there would be no live witnesses or jury; counsel could offer documentary and physical evidence, and depositions by stipulation, argument, and offers of proof; and, if the plaintiff won, the judge would set the damage award between $22,500 and $52,500.

During the trial, the appellant proffered evidence of the respondent’s tax returns. Appellant asserted such evidence would show whether the respondent had received investment tax credits and/or depreciation allowances for the purchase of the computer system. Appellant’s theory was that if received, such tax benefits should be considered as an offset to any compensatory damage award to which the respondent might be entitled thus lowering appellant’s liability. The court sustained respondent’s objection to the proffered evidence as irrelevant.

Appellant filed his notice of appeal on December 24, 1987. Briefing, however, was not completed until August 14, 1990.

Appellant’s sole contention is that the trial court erred by excluding the appellant’s proffered evidence of respondent’s tax returns as irrelevant.

*1538 Respondent requests sanctions, claiming intentional delay and a frivolous appeal.

Discussion

I. Standard of Review

In determining the admissibility of evidence, a trial court starts with the proposition “all relevant evidence is admissible.” (Evid. Code, § 351.) Relevant evidence is all evidence “including evidence relevant to the credibility of a witness or hearsay declarant, having any tendency in reason to prove or disprove any disputed fact that is of consequence to the determination of the action.” (Evid. Code, § 210.) In applying this standard, the court is given wide discretion to determine relevance under the code. (People v. Green (1980) 27 Cal.3d 1, 19 [164 Cal.Rptr. 1, 609 P.2d 468].) The appellate court should reverse only when a prejudicial abuse of discretion has occurred. (Mission Imports Inc. v. Superior Court (1982) 31 Cal.3d 921, 932 [184 Cal.Rptr. 296, 647 P.2d 1075].)

II. Introduction

Appellant asserts that by not considering the evidence of tax benefits which the respondent may have received, the trial court (1) applied the “collateral source rule” and (2) may have given the respondent a compensatory award which exceeded the limitations of the contract damage statutes.

Appellant’s first contention misconceives the issue; therefore, we will dismiss it before moving to the applicable law of this case.

III. The Trial Court Did Not Abuse its Discretion by Refusing to Grant Appellant’s Request for Collateral Source Credit.

In Anheuser-Busch, Inc. v. Stanley (1946) 28 Cal.2d 347 [170 P.2d 448, 166 A.L.R. 198] our Supreme Court stated the collateral source rule: “Where a person suffers personal injury or property damage by reason of the wrongful act of another, an action against the wrongdoer for the damages suffered is not precluded nor is the amount of damages reduced by the receipt by him of payment for his loss from a source wholly independent of the wrongdoer.” (Id. at p. 349.) Appellant is correct that since AnheuserBusch, supra, the rule has been applied in tort, but has never been extended to breach of contract. (Post, pp. 1539-1540; Patent v. Simpson (1967) 256 Cal.App.2d 506, 511 [64 Cal.Rptr. 187].) Therefore, appellant implies that *1539 by applying the collateral source rule in a breach of contract action, the trial court committed procedural error.

Appellant is mistaken, however, because the court did not apply the rule. The court stated its understanding of the law: defendants are not entitled to “collateral source credit for how a party [the plaintiff] handles their income tax.” (Italics added.)

A court’s decision not to grant a defendant collateral source credit does not rely on the same legal basis as a decision to apply the collateral source rule. As stated above, we agree with appellant: the collateral source rule has never been extended to breach of contract. However, not granting the appellant collateral source credit was well within the judge’s discretion. In California, there is no statutory or common law precedent in a breach of contract action to give defendants collateral source credit for income tax benefits which plaintiffs may have received. (Post, this page and p. 1540.)

IV. Tax Benefits Should Not be Considered as a Mitigating Factor in Compensatory Damage Calculations for Breach of Contract.

We begin with the California damage statutes for breach of contract. Civil Code section 3300 expressly allows plaintiffs to recover damages equivalent to the full detriment they suffer from the breach. 2 Civil Code section 3358 limits the award to no more than the “full performance” of the contract. 3 However, as instructive as the statutes are in defining what the upper and lower limits of the damage awards should be, they are silent as to how precise the calculations for those awards must be. They are equally silent on the question of whether tax benefits should be considered. 4 Furthermore, we find nothing in the legislative history of the statutes which *1540 reveals whether the Legislature intended tax benefits should or should not be considered in the trier’s determination of damage awards. 5

Similarly, California case law does not specifically address our issue, although for causes of action other than breach of contract, the courts have consistently prohibited evidence of tax considerations in the trier’s determination of damage awards. (Danzig v. Grynberg (1984) 161 Cal.App.3d 1128, 1139-1140 [208 Cal.Rptr. 336] [tax benefits which limited partners received from investment irrelevant in restitution for fraud]; Rodriguez v. McDonnell Douglas Corp. (1978) 87 Cal.App.3d 626, 665-668 [151 Cal.Rptr. 399] [income tax consequences of lost future income in determining award for personal injury irrelevant]; City of Los Angeles v. Tilem (1983) 142 Cal.App.3d 694, 707-708 [191 Cal.Rptr. 229] [tax benefits to property owner from inverse condemnation award irrelevant].)

There are few cases from other states on our issue; however, all of the cases we found hold tax consequences should not be considered as offsets for defendants in breach of contract actions. 6 In Billings Clinic v. Peat Marwick Main Co. (1990) 244 Mont. 324 [797 P.2d 899] the Montana Supreme Court held against tax benefits as offsets because: (1) if the defendant had not breached the contract, the plaintiffs would have been entitled to receive tax benefits as a benefit of the bargain, (2) considering tax benefits as offsets would be unduly speculative, and (3) “. . . the rule that damages must be reasonably certain in their nature and origin applies with equal force to claimed offsets to damages.” (Id. p. 913, citing Anderson v. Burlington Northern Inc. (1985) 218 Mont. 456 [709 P.2d 641, 648]; Tribby v. Northwestern Bank of Great Falls (1985) 217 Mont. 196 [704 P.2d 409, 417]; Bracy v. Great Northern Railway Company (1959) 136 Mont. 65 [343 P.2d 848, 853].) In Powers v. Powers (Tex.App. 1986) 714 S.W.2d 384 the Texas Appellate Court stated: “We are unaware of any principle or authority which would allow an offset to the party who has breached a contract for a ‘tax savings’ the non-breaching party ‘realized’ as a result of the breach.” (Id. p. 389.) We agree with the holdings of both courts.

There are several grounds for our conclusion tax consequences should not be considered as a mitigating factor in compensatory damage calculations in breach of contract cases. Those grounds are: (A) the federal “tax benefit rule” cancels out most windfalls to plaintiffs; (B) estimating tax *1541 consequences is not useful because it is speculative, time-consuming, and confusing; (C) public policy is better served by having defendants pay the full amount of the damages they have caused by breaching their contracts.

A. The Tax Benefit Rule Effectively Prevents Multiple Recoveries.

In the vast majority of cases, the federal tax benefit rule prevents plaintiffs from reaping multiple recoveries; therefore, it is the first reason we decide evidence of tax benefits should not be used as an offset against a defendant’s liability for compensatory damages in breach of contract actions. 7 The rule is a “. . . judicially developed principle codified, in part . . .” in the Internal Revenue Code, 26 United States Code Annotated, section 111. (See Mertens, Law of Federal Income Taxation (1990) § 7.74, p. 175.) 8

The tax benefit rule allows the government to recapture past tax benefits awarded to a taxpayer if in a later year an event occurs which changes the basis of, or is “. . . fundamentally inconsistent with[,] the premise upon which the deduction was initially based.” 9 (Hillsboro National Bank v. Commissioner (1983) 460 U.S. 370, 383 [75 L.Ed.2d 130, 146, 103 S.Ct. 1134].) The United States Supreme Court’s test for a “fundamentally inconsistent” event is “. . . if that event had occurred within the same taxable year, it would have foreclosed the deduction.” (Id. pp. 383-384 [75 L.Ed.2d at p. 146].)

It is not within this court’s jurisdiction to decide whether the tax benefit rule would apply to respondent in this circumstance: that is a matter for the *1542 Internal Revenue Service and the federal court system. Nonetheless, we cannot ignore the probability that when respondent receives his damage award, the rule will most likely cancel out any tax benefits which he may have received due to his purchase of the equipment from appellant. The receipt of the compensatory damage award may well be found to be fundamentally inconsistent with the previously awarded investment tax credit because if it had occurred within the year the credit was taken, the credit would not have been allowed. Therefore, under the tax benefit rule, respondent will probably be required to pay income tax on a portion of the damage award in the year it is received.

*1541 “Recovery of tax benefit items. Credits . . .

If-(A) a credit was allowable with respect to any amount for any prior taxable year, and (B) during the taxable year there is a downward price adjustment or similar adjustment, the tax imposed by this chapter for the taxable year shall be increased by the amount of the credit attributable to the adjustment.”

*1542 In Danzig, supra, 161 Cal.App.3d 1128, the First District cited the tax benefit rule as its rationale for ruling evidence of tax consequences is irrelevant to the setting of damages in fraud restitution actions. (Danzig, supra, 161 Cal.App.3d 1128, 1139; accord Burgess v. Premier Corp. (9th Cir. 1984) 727 F.2d 826, 837-838 and Cereal Byproducts Company v. Hall (1958) 16 Ill.App.2d 79 [147 N.E.2d 383].)

Similarly, the United States Supreme Court in Randall v. Loftsgaarden (1986) 478 U.S. 647 [92 L.Ed.2d 525, 106 S.Ct. 3143] (limited partners sued promoter for security fraud under § 12(2) of the Securities Act of 1933) held that tax benefits should not be considered in the determination of a damage award. The court explained that under the tax benefit rule any recovery by the investors would be taxable as ordinary income. “Any residual gains to plaintiffs thus emerge more as a function of the operation of the Internal Revenue Code’s complex provisions than of an unduly generous standard for defrauded investors.” (Id. at p. 664 [92 L.Ed.2d at p. 542].)

In Randall, supra, the Supreme Court resolved a conflict between the circuits: the Ninth had been ignoring tax benefits in its determinations of damage awards under the Securities Act and the court preferred its approach over the Second and Eighth which had been interpreting the act to mean that rescissionary damage awards should be reduced by the amount of the federal income tax benefits received from the original investment. (See Burgess v. Premier Corp. (9th Cir. 1984) 727 F.2d 826, 838; Austin v. Loftsgaarden (8th Cir. 1982) 675 F.2d 168 [which became Randall on appeal and which was specifically reversed]; Hayden v. McDonald (8th Cir. 1984) 742 F.2d 423, 439; Salcer v. Envicon Equities (2d Cir. 1984) 744 F.2d 935.)

Although we do not agree with the overruled policy of the Second and Eighth Circuits, their analysis of the issue is insightful. In criticizing the Ninth Circuit for ignoring tax consequences, the Hayden court stated, “These courts assume that past tax benefits and future tax consequences of *1543 recovery (under the tax benefit rule) would thus automatically cancel each other out.... [Cites] . . . The cases taking this position fail to recognize that the tax benefit rule achieves only a rough facsimile of the actual financial effects of deductions allowed in subsequent years.” (Hayden v. McDonald, supra, 742 F.2d 423, 440.)

The Hayden court’s view that the tax benefit rule yields only a “rough facsimile” is an accurate characterization because in Alice Phelan Sullivan Corporation v. United States (1967) 180 Ct.Cl. 659 [381 F.2d 399], the Court of Claims held “. . . the taxpayer must include in its income for the year of recovery the full amounts earlier deducted and must pay tax at the rate prevailing in the year of recovery.” (See Mertens, Law of Federal Income Taxation, supra, § 7.81, p. 199.)

Consequently, changes in the taxation rates may allow a recovery which arguably is not precisely consonant with the California damage statutes. If the tax rate prevailing in the year of recovery were lower than in the year the deduction was taken, a taxpayer who receives a damage award could receive somewhat more compensation (counting his net tax benefits) than the full performance which is allowed under Civil Code section 3358. Conversely, if the tax rate were higher, the taxpayer could receive less than what the Legislature intended by enacting Civil Code sections 3300 and 3358.

We concede the tax benefit rule does not work perfectly in all situations. We do not believe, however, that the few instances in which the rule does not precisely cancel out a plaintiff’s tax gain result in violations of the contract damage rules. As discussed, the breach of contract damage statutes do not state how precise the trier must be in making a determination of damages, and California courts have not interpreted the statutes as requiring determinations to the exact dollar. 10

In interpreting another ambiguous damage limitation law, section 28 of the federal Securities Exchange Act of 1934, the United States Supreme Court in Randall v. Loftsgaarden, supra, 478 U.S. 647 stated: “We think that § 28(a) [limiting recovery to “net economic loss”] cannot be fairly read to require such a full-scale inquiry into a defrauded investor’s dealing with the tax collector lest the investor escape with anything more than his ‘net economic loss.’ ” (Id. at p. 665 [92 L.Ed.2d at pp. 542-543].) We agree with the approach in Randall, and do not think Civil Code section 3358’s limita *1544 tion of full performance means the trier must scrutinize a plaintiff’s tax returns in order to determine the exact amount of a compensatory damage award.

B. Estimating Tax Benefits Is Speculative, Time-consuming, and Confusing.

Even if there were no tax benefit rule, there are other reasons for not allowing the trier to consider tax benefits in its damage determinations. Foremost among them is the necessarily complex and speculative nature of inquiring into tax consequences. As the United States Supreme Court stated in Randall, “The Court of Appeals’ elaborate method for calculating damages and interest so as to offset tax benefits supplies an additional reason for rejecting its tax benefit offset rule .... [T]here are formidable difficulties in predicting the ultimate treatment of the investor’s claimed tax benefits, whether or not an audit has commenced and . . . the burdens associated with reconstruction of the investor’s tax history . . . are substantial.” (478 U.S. at pp. 664-665 [92 L.Ed.2d at p. 542].)

Similarly, in Bracy v. Great Northern Railway Company, supra, 136 Mont. 65, the Montana Supreme Court observed: “The tax liability of today is no criterion of what it may be tomorrow. It has a faculty of constantly changing in the face of vocal threats of reduction usually made on the eve of an election.” (Id. p. 74.)

We agree with the United States and Montana Supreme Courts. Experience teaches us it is very complicated and speculative to predict: (1) what portion of a damage award will actually be paid; (2) the year or years when a damage award will actually be received; (3) what the prevailing tax rate will be during the year(s) of receipt; and (4) how a damage award will be construed under the continuously changing Internal Revenue Code. We are concerned even the most sophisticated attempts to make a prediction about tax consequences would result in only guesswork which has little probative value.

Furthermore, although Evidence Code section 352 is not controlling in the present case, and its application is not under review, we think a fair application of the rule prohibits admission of tax consequences. By enacting Evidence Code section 352, the Legislature gave the court the means to facilitate judicial economy. This section expressly gives the trial court discretion to exclude evidence “. . . if its probative value is substantially outweighed by the probability that its admission will (a) necessitate undue consumption of time or (b) create substantial danger of. . . confusing the issues . . . .” (Evid. Code, § 352.)

*1545 In applying the criteria set forth in Evidence Code section 352 to evidence of tax consequences, we observe the following: (1) such evidence would only be useful in those few instances when the tax rate changed; and (2) even the most accurate predictions of the tax consequences would result in mere approximations. (See ante, pp.1541-1543.) On the other side of the scale, trial judges and juries would be subjected to complicated, time consuming and confusing processes which would be required to admit and consider such minimally relevant evidence. Consequently, it appears to us any reasonable trial judge would rule against the admission of such evidence under Evidence Code section 352, and this court would uphold an exercise of discretion in that direction. At the trial the court stated if it had not ruled the evidence inadmissible on relevancy grounds, it would have ruled it inadmissible under Evidence Code section 352. We agree and would affirm that exercise of discretion.

C. Public Policy Is Better Served if Defendants Who Have Breached Contracts Are Responsible for the Full Amount of Compensatory Damages.

Even if evidence of tax consequences were admissible under Evidence Code section 352, we would still rule against the appellant for reasons of public policy:

(1) There is deterrent value against breaching a contract if contracting parties know they must compensate the other party in full, without being able to subtract any tax benefits the other party might receive. As interested participants in an already overloaded court system, we, as well as those who are counting on full performance of their contracts, have a vested stake in discouraging others from breaching those contracts.

(2) Even if there were no tax benefit rule or the Internal Revenue Service decided against applying the rule, we would reject appellant’s theory. We will assume arguendo respondent had received tax benefits, his damage award will not be subject to the tax benefit rule, and he will be receiving a windfall recovery (i.e., the tax benefits he received added to his damage award of $40,000 will yield him a recovery greater than the $40,000 in damages he actually suffered). Under these hypothetical circumstances appellant would have us deduct respondent’s tax benefits from appellant’s $40,000 damage liability rather than allow respondent to reap the windfall. Under appellant’s scheme, therefore, we should in effect take away respondent’s tax benefits and credit them to appellant.

We disagree: tax benefits should be received by only those for whom the government intended them. In the present case, if respondent received an *1546 investment tax credit for purchasing the computer system, it was because the government was encouraging the purchase of such systems by conferring credits. For this court to rule such a subsidy should now in effect be taken from respondent and credited to appellant would be at odds with the purpose of the credit; it would also be tantamount to awarding appellant government funding to subsidize his breach of contract. Since this court does not believe taxpayers should subsidize those who have not fulfilled their contractual obligations, we find little merit in appellant’s theory.

(3) If the hypothetical scenario (above) existed we would not be enthusiastic about allowing respondent to receive a multiple recovery; however, as the United States Supreme Court in Randall v. Loftsgaarden, supra, 478 U.S. 647 stated: “ ‘. . . [I]t is more appropriate to give the defrauded party the benefit even of windfalls than to let the fraudulent party keep them.’ ” (Id. at p. 663 [92 L.Ed.2d at p. 541] citing Janigan v. Taylor (1st Cir. 1965) 344 F.2d 781, 786.) Similarly, in actions for breach of contract, this court favors parties who honor their promises, not those who breach them.

V. We Deny Respondent’s Request for Sanctions Because the Appeal Was Not Totally and Completely Without Merit, and the Delay Was Not Intentional.

We decline respondent’s request to impose sanctions against appellant for intentional delay and a frivolous appeal. In the case of In re Marriage of Flaherty (1982) 31 Cal.3d. 637 [183 Cal.Rptr. 508, 646 P.2d 179] our Supreme Court instructed “[A]n appeal should be held to be frivolous only when it is prosecuted for an improper motive—to harass the respondent or delay the effect of an adverse judgment—or when it indisputably has no merit—when any reasonable attorney would agree that the appeal is totally and completely without merit.” (Id. at p. 650.)

Despite ruling against him, we believe appellant’s appeal was not totally and completely without merit. Since the present case is one of first impression in California, appellant could not have known how this court would rule. (Ante, pp. 1539-1540.) With no statute or case law precedent to consult for explanation, appellant was justified in requesting a clarification of the law.

Our review of the record indicates that although appellant had great difficulty meeting court deadlines, there were many other factors not in appellant’s control which were even more instrumental in slowing the prog *1547 ress of this appeal. Therefore, we conclude appellant did not intentionally delay the appeal process in order to reap unwarranted benefit.

Disposition

The judgment is affirmed. Appellant to pay respondent’s costs on appeal.

Lillie, P. J., and Woods (Fred), J., concurred.

We are not addressing the issues of whether a plaintiff may recover tax benefits which are lost due to a defendant’s breach, or whether damage awards for breach of contract should be increased by the amount of income tax plaintiffs expect to pay on their award. (See Annot., 50 A.L.R.4th 452.) Similarly, the Evidence Code does not instruct whether evidence of respondent’s tax benefits would or would not be privileged; we do not reach the issue in this opinion.

Civil Code section 3300 provides:

“For the breach of an obligation arising from contract, the measure of damages ... is the amount which will compensate the party aggrieved for all the detriment proximately caused thereby, or which . . . would be likely to result therefrom.”

Civil Code section 3358 provides:

“Except as previously provided by statute, no person can recover a greater amount in damages for the breach of an obligation, than he could have gained by the full performance thereof on both sides.

Section 3358 is similar to Corbin’s general rule: “[C]are must be taken not... to permit the plaintiff to receive double or excessive compensation for his injury.” (See 5 Corbin on Contracts (1964) § 1036, p. 215.)

The Restatement Second of Contracts is also silent on the issue of whether tax benefits should be considered as offsets to a defendant’s damage award. Our issue does, however, fall under the general provisions of section 351: “A court may limit damages for foreseeable loss . . . if it concludes that in the circumstances justice so requires in order to avoid disproportionate compensation.” (Rest.2d Contracts, § 351, p. 135.)

Civil Code section 3300 was enacted in 1872, before the Sixteenth Amendment to the Constitution established the federal income tax. Also enacted in 1872, Civil Code section 3358 was amended in 1977 to allow for liquidated damages, but the tax consequences of damage awards were not addressed by the Legislature.

Seepost, pages 1541-1543, for why federal courts do not consider tax credits as a mitigating factor in securities fraud cases.

We need not discuss the issue as it pertains to California income tax because the federal tax benefit rule also accounts for prior deductions under California tax law. This is because California refers to the federal law for its: (1) definitions of “gross income,” “adjusted gross income,” and “taxable income;” (2) determinations of which items are included and excluded from gross income; and (3) determinations of “itemized deductions.” (See 9 Witkin, Summary of Cal. Law (9th ed. 1989) Taxation, § 287, p. 344.)

Not all property or contracts fall under the tax benefit rule; there are other Internal Revenue Code sections which guarantee that the government is able to recover tax benefits when the “basis” for those benefits changes in a later year. (Russell v. French & Associates, Inc. (Tex.Ct.App. 1986) 709 S.W.2d 312, 315 [a tax benefit received in a previous year on an interest in oil, gas, geothermal, or other mineral properties will be cancelled out by an adjustment required by 26 U.S.C.A. § 1254 if the contract for the purchase of those properties is rescinded].)

26 United States Code Annotated, section 111 states in relevant part:

(See ante, pp. 1539-1540; Arques v. National Superior Co. (1945) 67 Cal.App.2d 763, 779 [155 P.2d 643] [“In an action for a breach of contract, the fact that the amount of damage may not be susceptible of exact proof does not bar recovery.’’] citing Long Beach Drug Co. v. United Drug Co. (1939) 13 Cal.2d 158 [88 P.2d 698].)