Opinion

Starr Starr appeals from the judgment entered after the family law court found that the house he bought in his name only while married to former wife Martha Starr was community property and ordered him to convey the property to them both as tenants in common. The evidence shows that Martha quitclaimed her interest in the house based on Ron’s promise to put her on title after the purchase was completed, but that Ron failed to do so. As a result, the evidence supports a finding that the house was community property based on Ron’s violation of his fiduciary duties to Martha. We also conclude that the trial court properly valued the community and separate property interests in the house, and did not err in denying Ron’s request to refund his overpayment of child support credits. We therefore affirm the judgment.

FACTS AND PROCEDURAL HISTORY

In late 1996, Ron Starr bought a house in Glendora, taking title in his name only as his separate property even though he was then married to Martha Starr. Ron filed for divorce in April 2004. In his petition, signed under penalty of perjury, he listed the house as community property, but sought the return of his separate property contributions to the property. By the time of trial, however, Ron contended the house was his separate property.

Ron testified that the house was bought in his name only because the $50,000 downpayment came from his separate property funds, and he and Martha intended all along that the house would be his separate property. In accord with that plan, Martha quitclaimed her interest in the house before escrow closed. Property taxes and mortgage payments came from community property earnings, Ron testified.

Under Family Code section 721, Ron had the burden of proving that the quitclaim transaction satisfied his fiduciary duties to Martha. She testified that because of her poor credit history, the lender recommended she agree to the quitclaim so she and Ron could qualify for a better interest rate. The loan broker told Martha and Ron they could add Martha back onto the title by way of a quitclaim deed within 45 days of the close of escrow. Martha had a discussion with Ron about adding her onto the title, and he said he would do that. Martha said she and Ron jointly offered to buy the house, and that the deed to Ron was mailed to them both after it had been recorded. Although Ron never added Martha onto the title, she never worried about it because “He’s my husband. I just don’t . . . mistrust him. You know, it was our house.” She signed the quitclaim deed freely and voluntarily.

Ron was impeached with his deposition testimony, where he said title was taken in his name in order to facilitate the financing. When asked on cross-examination about the statement on his divorce petition that the house was community property, Ron said he could not recall whether his former lawyer went over his assets with him before signing the petition, and that he probably did not read it before signing.

The trial court found that the house was community property, but that Ron was entitled to reimbursement of the $50,000 downpayment from his separate property funds. Ron was ordered to convey the house to himself and Martha as tenants in common. In its statement of decision, the court said the “controlling cases on the issue” were In re Marriage of Benson (2005) 36 Cal.4th 1096 [ 32 Cal.Rptr.3d 471 , 116 P.3d 1152 ] (Benson), In re Marriage of Mathews (2005) 133 Cal.App.4th 624 [ 35 Cal.Rptr.3d 1 ] (Mathews), In re Marriage of Delaney (2003) 111 Cal.App.4th 991 [ 4 Cal.Rptr.3d 378 ] (Delaney), and In re Marriage of Haines (1995) 33 Cal.App.4th 277 [ 39 Cal.Rptr.2d 673 ] (Haines). In a separate paragraph, the trial court found that Ron did not meet his burden of proof that Martha’s quitclaim deed was signed “freely and voluntarily. The reason [Martha] did not sign the quitclaim deed freely and voluntarily was because the intent of the lender controlled title to the [house] when the lender suggested that [Martha’s] name be left off of the mortgage for the purposes of financing, and [Martha] agreed to execute the quitclaim deed based on the lender’s suggestion.”

Ron contends the trial court erred because it relied on the “lender’s intent” theory, which is applicable only to determining whether loan proceeds obtained during marriage are community or separate property. Instead, according to Ron, the court should have applied the reasoning of the factually similar Mathews, supra, 133 Cal.App.4th 624 , and found that he satisfied his fiduciary obligations to Martha based on her testimony that she signed the quitclaim deed freely and voluntarily.

DISCUSSION

1. Family Code Section 721

Although spouses may enter transactions with each other (Fam. Code, § 721, subd. (a)), such transactions “are subject to the general rules governing fiduciary relationships which control the actions of persons occupying confidential relations with each other. This confidential relationship imposes a duty of the highest good faith and fair dealing on each spouse, and neither shall take any unfair advantage of the other. This confidential relationship is a fiduciary relationship subject to the same rights and duties of’ unmarried business partners, including the right of access to records and information concerning their transactions. (Fam. Code, § 721, subd. (b).)

Because of this, our courts have long held that when an interspousal transaction advantages one spouse, public policy considerations create a presumption that the transaction was the result of undue influence. ( Haines, supra, 33 Cal.App.4th at pp. 293-294.) A spouse who gained an advantage from a transaction with the other spouse can overcome that presumption by a preponderance of the evidence. ( Mathews, supra, 133 Cal.App.4th at pp. 631-632.)

2. The Haines Decision

In Haines, a wife who quitclaimed her interest in the house she jointly owned with her husband sought to invalidate the deed during their divorce proceedings because she was coerced into signing it. The wife testified that she and her husband had several arguments about signing the deed as their marriage deteriorated. She claimed the husband ranted and raved, pulled her hair, and threw water in her face during one of these arguments. Later, the husband agreed to cosign a loan for the wife so she could buy herself a car that she would need once she was on her own. While the husband was driving the wife to her credit union to cosign the loan, he told her he would not do so unless she agreed to the quitclaim deed. She signed the deed because she felt she had no alternative.

Evidence Code section 662 creates a presumption that title is actually held as described in a deed. The trial court found that the wife failed to meet her burden of rebutting that presumption by clear and convincing evidence, as required by that provision, but found she would have satisfied it if the preponderance of the evidence standard of proof had applied. The Haines court quoted Brison v. Brison (1888) 75 Cal. 525, 529 [ 17 P. 689 ] (Brison T) for the proposition that when a spouse gained an advantage from a transaction with the other spouse, “ ‘[t]he law, from considerations of public policy, presumes such transactions to have been induced by undue influence.’ ” (Haines, supra, 33 Cal.App.4th at p. 293 .) When that presumption arose, it trumped the competing presumption created by Evidence Code section 662. (Haines, at pp. 297, 299-301.) Therefore, the husband had to show that the deed “ ‘was freely and voluntarily made, and with a full knowledge of all the facts, and with a complete understanding of the effect of the transfer.’ ” (Id. at p. 296, quoting Brown v. Canadian Indus. Alcohol Co. (1930) 209 Cal. 596, 598 [ 289 P. 613 ].) Because the trial court found the wife met the lesser standard of proof applicable to the section 721 presumption, it reversed the judgment and held that the quitclaim deed was invalid. (Haines, at p. 302.)

3. The Mathews Decision

The facts in Mathews are similar to this case. While a husband and wife were in the process of buying a house, the wife quitclaimed her interest in the house to the husband in order to obtain a better interest rate. Title to the house was taken in the husband’s name alone. The wife knew title was taken in that manner, but believed she would be added to the title later on. The wife contested the validity of the quitclaim deed, primarily on the basis that she was a native of Japan and did not speak English well enough to fully understand what she was doing. The trial court refused to apply section 721’s presumption of undue influence and awarded the house to the husband as his separate property. The Court of Appeal held that the trial court erred by refusing to apply the presumption, because the husband clearly gained an advantage when the wife quitclaimed her interest to him. (Mathews, supra, 133 Cal.App.4th at p. 630 .)

Citing to Haines, supra, 33 Cal.App.4th at page 296 , the Mathews court held that the husband had to prove the quitclaim deed was freely and voluntarily made, with full knowledge of all the facts and a complete understanding of its effects. (Mathews, supra, 133 Cal.App.4th at p. 631 .) The trial court’s error was harmless, the Mathews court held, because the evidence supported the trial court’s finding that the wife freely and voluntarily quitclaimed her interest in the house with full knowledge of the facts. This included evidence that: the wife asked questions when she did not understand something, but asked none when she signed the quitclaim deed; the husband put no pressure on her to sign; and she did so in order to get better financing; and completion of the purchase did not depend on her signing the quitclaim deed. Although the wife was a native of Japan, evidence that she was fully fluent in English, handled her own separate finances as well as their joint finances, and admitted knowing her name was not on the title but “assumed it would be added later,” led the court to conclude the quitclaim deed was “valid and executed freely and voluntarily in good faith.” As a result, the husband “rebutted the presumption of undue influence by a preponderance of the evidence.” ( Mathews, supra, 133 Cal.App.4th at pp. 631-632.)

4. Mathews Is Not Applicable; Instead Ron’s Failure to Add Martha onto the Title Was a Breach of His Fiduciary Duty

It is easy to see why Ron relies on Mathews . The factual setting seems virtually identical to this case, with the added bonus of Martha’s testimony that she signed the quitclaim deed freely and voluntarily. There is a critical— and we believe fatal—distinction, however. In Mathews , the wife said she merely assumed she would be added onto the title after escrow closed, while Martha testified that Ron told her he would do so. The importance of this distinction is tied up in both section 721 and its statutory predecessor, and the sometimes confusing use of the term “undue influence” by decisions interpreting those provisions.

Section 721 never mentions undue influence. Instead, it states that spouses are in a confidential and fiduciary relationship and have a duty to each other of the highest good faith and fair dealing. Its predecessor, Civil Code former section 158 (Cal. Law Revision Com. com., 29C West’s Ann. Fam. Code (2004) foil. § 721, p. 267) was substantially similar.

Despite that omission, the court in Haines, supra, 33 Cal.App.4th 277 , relying on Brison I, supra, 75 Cal. at page 529 , noted that an interspousal transaction that benefits one of the spouses creates a presumption of undue influence, requiring the husband who obtained his wife’s quitclaim deed to the family home to show that the deed was freely and voluntarily made. Mathews, supra, 133 Cal.App.4th at page 630 , cited Haines for the same proposition. The trial court’s statement of decision in this case made findings concerning whether Martha’s quitclaim deed was freely and voluntarily made, and Ron has understandably focused his appellate arguments on the concept of undue influence as expressed in Mathews .

Undue influence is a contract defense based on the notion of coercive persuasion. Its hallmark is high pressure that works on mental, moral, or emotional weakness, and it is sometimes referred to as overpersuasion. (Odorizzi v. Bloomfield School Dist. (1966) 246 Cal.App.2d 123, 130 [ 54 Cal.Rptr. 533 ].) Undue influence is statutorily defined as taking unfair advantage of another’s weakness of mind (Civ. Code, § 1575, subd. 2), or taking a grossly oppressive or unfair advantage of another’s necessity or distress (Civ. Code, § 1575, subd. 3). Haines and Mathews appear to fall into these categories. ( Mathews, supra, 133 Cal.App.4th at pp. 631-632 [husband overcame presumption of undue influence because the evidence showed he did not pressure wife, who understood what she was signing]; Haines, supra, 33 Cal.App.4th at pp. 283-284 [husband berated and assaulted wife, and refused to cosign a car loan unless she agreed to quitclaim her interest in their house].)

However, there is another type of conduct that amounts to undue influence: the use of confidence or authority to obtain an unfair advantage. (Civ. Code, § 1575, subd. 1.) This is triggered by one party’s breach of a confidential relationship. (O’Neil v. Spillane (1975) 45 Cal.App.3d 147, 152-153 [ 119 Cal.Rptr. 245 ].) It is also the type of conduct at issue in Brison I, supra, 75 Cal. 525 , which, as we next discuss, explains not only how “undue influence” became shorthand for conduct that violates section 721, but why the evidence in this case supports the judgment.

In Brison I , a husband was about to embark on a long and potentially perilous business trip. Based on the wife’s promise that she would reconvey to him upon request, the husband deeded real property to her so she could avoid going through probate if he died. When he returned and the wife refused to deed back the property, he sued to compel a reconveyance. The Supreme Court reversed the trial court’s order sustaining a demurrer to the complaint. The complaint alleged that under Civil Code former section 158, the parties were in a confidential relationship, and the husband was induced to make the deed based on his confidence in her and her promise to reconvey. “The betrayal of such confidence is constructively fraudulent, and gives rise to a constructive trust. This is independent of any element of actual fraud. [Citation.] The law, from considerations of public policy, presumes such transactions to have been induced by undue influence. [Citations.]” (Brison I, supra, 75 Cal. at p. 529 .)

Upon remand from Brison I , the case went to trial, resulting in a judgment for the husband, and another appeal. (Brison v. Brison (1891) 90 Cal. 323 [ 27 P. 186 ] (Brison II).) The wife appealed from an order denying her motion for a new trial, and the Supreme Court affirmed, holding that the evidence justified that order. In accord with his complaint, the husband testified that he deeded the property to his wife based on her promise to reconvey, and that he intended her to have the property only if he died. The Supreme Court held that the evidence of the wife’s “subsequent refusal to reconvey was not merely the breach of an agreement, but was the betrayal of a confidence, and the violation of a trust, constituting a constructive fraud, which a court of equity will remedy. The influence which the law presumes to have been exercised by one spouse over the other is not an influence caused by any act of persuasion or importunity, but is that influence which is superinduced by the relation between them, and generated in the mind of the one by the confiding trust which he has in the devotion and fidelity of the other. Such influence the law presumes to have been undue, whenever this confidence is subsequently violated or abused.” (Brison II, supra, at p. 336, citing to Civ. Code, § 1575, subd. 1.)

Brison I and II are significant for three reasons. First, they announced the overarching principle that constructive fraud due to breach of a confidential relationship amounts to undue influence, terminology that was adopted by other courts. Second, they differentiated such constructive fraud from the other forms of undue influence based on acts of coercion or overpersuasion. Third, they established a paradigm of constructive fraud arising from one spouse’s conveyance of property to the other spouse based on an unfulfilled promise by the other spouse to reconvey. This fact pattern has been applied by our courts many times in cases involving spouses and other persons in confidential relationships. (See, e.g., Alaniz v. Casenave (1891) 91 Cal. 41, 46 [ 27 P. 521 ] [in action to reconvey deeds conveyed to trusted family member, court held that if a deed is obtained without consideration by way of an oral promise to reconvey in a transaction between those in a confidential relationship, the breach of promise is constructive fraud]; Sparks v. Sparks (1950) 101 Cal.App.2d 129, 135 [ 225 P.2d 238 ] [in action to quiet title and void deed from father and one son to other son based on breach of fiduciary duties, judgment affirmed; undue influence is a species of constructive fraud and depends on the facts and circumstances of each case]; Holmes v. Holmes (1950) 98 Cal.App.2d 536, 538 [ 220 P.2d 603 ] [affirming judgment for woman who sued man with whom she lived as husband and wife when man obtained deed to plaintiff’s restaurant on promise to invest his own funds and work at the restaurant; because they behaved as a married couple, the same confidential relationship arose, and the man’s breaches of promise were constructive fraud]; Hilton v. Hilton (1921) 54 Cal.App. 142, 155 [ 201 P. 337 ] [invalidating deed by wife to husband as part of divorce settlement on ground of undue influence; equating undue influence with lack of free will in entering transaction].)

Perhaps most notable of these for our purposes is Jones v. Jones (1903) 140 Cal. 587 [ 74 P. 143 ] (Jones), where a wife conveyed land to her husband on the advice of a lawyer who told them the transfer was required in order to bring an action to eject a tenant in possession of the land. Instead, the husband conveyed the land to a third party so the third party could bring the action, and the third party then claimed he was the true owner. The wife sued her husband and the third party. The trial court found those facts were true, but found that the husband had not acted fraudulently, but instead intended to carry out the plan to eject the tenant. The trial court entered judgment for the wife and enjoined the husband and the third party from making any claims to the property.

The Supreme Court affirmed, partly in reliance on Brison I, supra, 75 Cal. 525 . (Jones, supra, 140 Cal. at p. 590 .) Even if the attorney who advised the wife had been employed by her, the husband was not exonerated because, “by accepting the deed upon the statement made in his presence of the purposes for which he was to hold the land, [he] became a party to the transaction, and by implication promised to fulfill the purpose of the trust.” (Id. at p. 591.) As a result, the failure to fulfill this agreement was constructive fraud, allowing the wife to have the deed declared void. (Id. at p. 590.) In short, even when the suggestion to convey came from a third party adviser and no express promises were made by the husband, he impliedly promised to fulfill the conditions of the transfer, and the failure to do so was constructive fraud.

By substituting the Starrs’ loan broker for the lawyer in Jones , and adding in an express promise by Ron to essentially reconvey Martha’s quitclaimed interest in place of an implied promise, we believe Jones is applicable here. Viewing the evidence most favorably to the judgment, Martha and Ron were told by the lender they should have Ron take title in his name only, with Martha quitclaiming her interest in the house, so they could get a better interest rate. The lender said Ron could add Martha back onto the title after escrow closed, and Ron told Martha he would do that. Martha never questioned what happened because Ron was her husband and she trusted him. Ron declared under penalty of perjury in his divorce petition that the house was community property, but claimed he probably never read the declaration before signing it. This evidence falls squarely within the Brison-Jones paradigm and supports a finding that Ron’s failure to add Martha onto the title as promised was constructive fraud and undue influence, thereby breaching his fiduciary duty to Martha.

For the same reason, we hold that Mathews is not applicable here. While the Mathews court mentioned in passing that the wife “assumed” or “believed” she would be added onto the title, there is no indication that her husband ever promised that would happen. The Mathews court did not develop the point, and it played no part in that court’s analysis. Because the Brison-Jones fact pattern was not present in Mathews, and was not part of its decision, we hold that Mathews is not applicable here. We next consider whether the trial court’s statement of decision allows us to affirm based on such a finding.

5. The Statement of Decision Was Ambiguous

In nonjury trials, unless a statement of decision has been requested and rendered, we will presume that the trial court made all the factual findings necessary to support the judgment, so long as those implied findings are supported by substantial evidence. If a statement of decision is given, it provides us with the trial court’s reasoning on disputed issues and “is our touchstone to determine whether or not the trial court’s decision is supported by the facts and the law.” (Slavin v. Borinstein (1994) 25 Cal.App.4th 713, 718 [ 30 Cal.Rptr.2d 745 ].) Ron contends the trial court’s statement of decision shows that the trial court’s reasoning was flawed and that we should reverse the judgment because (1) of the four decisions listed in the trial court’s statement of decision, only one—Mathews—is legally and factually applicable; and (2) the trial court’s finding that the quitclaim deed was not signed freely and voluntarily is based on the lender’s intent doctrine, which is also inapplicable.

Implicit in Ron’s contentions is the notion that the statement of decision clearly and unambiguously shows the trial court erred. We disagree. Instead, we conclude that the statement of decision is ambiguous. Because the record does not show that Ron (or Martha) objected to the statement of decision, “whatever uncertainties may exist in the findings of the trial court are to be so resolved, if reasonably possible, as to support the judgment rather than defeat it [citation].” (Reinsch v. City of Los Angeles (1966) 243 Cal.App.2d 737 , 746 [ 52 Cal.Rptr. 613 ].) Therefore, we resolve all conflicts and ambiguities in the findings, and infer all logical and reasonable findings, in support of the judgment. (Associated Creditors’ Agency v. Dunning Floor Covering, Inc. (1968) 265 Cal.App.2d 558, 559 [ 71 Cal.Rptr. 494 ].)

The first “finding” Ron relies on is the paragraph in the statement of decision listing the four appellate decisions that the trial court believed were controlling. This is not a finding of fact, however. Instead it is immaterial surplusage that we may disregard. (Canadian Indem. Co. v. Motors Ins. Corp. (1964) 224 Cal.App.2d 8, 17-18 [ 36 Cal.Rptr. 159 ].)

The second finding Ron points to as the source of the trial court’s error is the statement that he did not meet his burden of showing by a preponderance of the evidence that Martha signed the quitclaim deed “freely and voluntarily. The reason [Martha] did not sign the quitclaim deed freely and voluntarily was because the intent of the lender controlled title to the [property] when the lender suggested that [Martha’s] name be left off of the mortgage for the purposes of financing, and [Martha] agreed to execute the quitclaim deed based on the lender’s suggestion.”

According to Ron, this shows the trial court was relying on a family law property characterization doctrine known as the lender’s intent theory, which applies to only the characterization of loan proceeds obtained during marriage. Ron correctly cites to In re Marriage of Grinius (1985) 166 Cal.App.3d 1179 [ 212 Cal.Rptr. 803 ] (Grinius) for the definition and applicability of that theory. (Id. at p. 1186 [the character of credit acquisitions during marriage is “ ‘determined according to the intent of the lender to rely upon’ ” either the separate property of one spouse or the community property of both].) He is wrong in contending that the trial court was unambiguously referring to that doctrine in its statement of decision.

Nothing in the record shows that this theory was ever mentioned at trial. Martha’s lawyer did not raise it during his closing argument, contending instead, but without elaboration, that the critical issue in regard to the house had to do with Ron’s fiduciary duties to her. Moreover, the Grinius court noted that it was not reaching any issues arising from the husband’s supposed breach of his fiduciary duty to his wife. (Grinius, supra, 166 Cal.App.3d at p. 1190, fn. 4 .) Ron does not explain how or why the trial court chose to rely on a theory that was not at issue and was without a doubt inapplicable to the issue before it: whether Ron breached his fiduciary duties to Martha. Furthermore, the trial court’s language leaves a gap between the lender’s “suggestion” and any conduct by Ron that influenced and affected Martha. Based on this, we conclude the finding is ambiguous. However, we also conclude the ambiguity can be resolved.

As we have seen, the failure to add Martha onto the title is constructive fraud under section 721, and constructive fraud is presumed to be undue influence, which means the transaction was not free and voluntary. When the trial court found Martha did not act freely and voluntarily, it necessarily found that she quitclaimed her interest in the house as the result of undue influence. (See In re Marriage of Dawley (1976) 17 Cal.3d 342 , 354—355 [ 131 Cal.Rptr. 3 , 551 P.2d 323 ] [when trial court found that wife signed antenuptial agreement freely and voluntarily, Supreme Court implied necessary finding that the transaction was not the result of undue influence].) Under Jones, supra, 140 Cal. 587 , the lender’s suggestion to have Martha quitclaim her interest in the house and have Ron add her to the title later, combined with Ron’s failure to fulfill his promise to do so, is constructive fraud amounting to undue influence. We believe that when the trial court referred to the lender’s suggestion as the source of the undue influence, it was attempting to make this connection. Once read this way, the finding of fact squares with both the applicable law and the facts of this case. We therefore construe the finding in that manner. (See Mintz v. Rowitz (1970) 13 Cal.App.3d 216, 227-228 [ 91 Cal.Rptr. 435 ] [in action by heirs to enforce agreement between spouses making mutual wills that the survivor would not alter agreed-upon estate distribution plan, trial court found for plaintiff heirs, but made inconsistent and irreconcilable finding that the wife, who was the first spouse to die, had fully understood the consequences of her will, which in fact did not include the agreed-upon distribution plan; under rule to construe ambiguous findings in support of judgment, appellate court affirmed, holding that “it seems quite clear that what the court meant to say was that” the wife understood her will to be a memorandum of the separate agreement].)

6. Valuation of Ron’s Separate Property Interest in the House

The trial court ordered that Ron be reimbursed his $50,000 separate property downpayment for the house, pursuant to Family Code section 2640. Ron contends this was error, because that section applies only to property taken in joint title during the marriage. That section provides that the spouse making a separate property downpayment is entitled to be reimbursed for his contribution. (Fam. Code, § 2640, subd. (b).) Ron contends this rule is subject to Family Code section 2580, which declares the Legislature’s findings and declarations in a part of the Family Law Code devoted to establishing a uniform method for determining the character of property acquired by spouses during marriage in joint title form.

We do not believe Family Code section 2580 somehow prevents a court from awarding separate property contributions to a spouse under these circumstances. Family Code section 2640, subdivision (b) states that reimbursement is awarded upon “division of the community estate,” and says nothing about the requirement that the property have first been acquired in joint title form. Furthermore, Ron’s contention is based on the assertion that the judgment did not change the fact that title was in his name as his separate property. He overlooks that the trial court ordered him to convey the property to himself and Martha as tenants in common, however. More fundamentally, we fail to see how Ron has been harmed by this award, when it in fact ensures that he is reimbursed for his separate property contribution. He does not contend that the $50,000 amount is in error, and does not suggest some other means of calculating the amount of his reimbursement. Because he has failed to show any harm from the trial court’s ruling, we will affirm.

7. Child Support Reimbursement

Ron and Martha have two children. In 2005, they stipulated to a DissoMaster-generated child support payment from Ron that was erroneously based on the entry of data showing they had four children. As a result, Ron overpaid his child support obligation by $3,112. In 2006, the court modified the support order to reflect the proper amount, and reserved for trial the issue of whether Ron was entitled to reimbursement of the overpayment. At trial, the court found that the then current support order was inadequate, increased the amount, and declined to reimburse Ron for his overpayments. Ron contends the trial court erred because it did not make the findings required to justify a child support award above the guideline amounts. Martha contends that under In re Marriage of Peet (1978) 84 Cal.App.3d 974 [ 149 Cal.Rptr. 108 ] (Peet) the court had discretion to deny the reimbursement claim.

Peet involved a spouse’s claim for reimbursement of voluntary overpayments, which the trial court granted. The Court of Appeal affirmed, holding that the trial court had discretion in such matters. We hold that Peet is applicable here. This is not a case where the trial court awarded child support above the guideline amounts in the first instance. Instead, based on a clerical error, the parties stipulated to an amount they believed was correct, and the court ordered support in that amount. Therefore, the requirements to deviate from the guideline amount (Fam. Code, § 4056) are not applicable. Given the trial court’s determination that support in the correct amount was inadequate, we hold that the court did not abuse its discretion by declining Ron’s request for reimbursement.

DISPOSITION

The judgment is affirmed. Respondent shall recover her appellate costs.

Flier, J., and Grimes, J., concurred.

We will refer to Martha Starr and Ron Starr by their first names. Ron was represented by counsel when he filed his divorce petition. At trial, he had a new lawyer, who asked the court to let Ron amend the petition to reflect his new contention. The court denied the request, but later said the pleadings were not conclusive and the issue was a matter of proof during the trial. We will refer to Family Code section 721 as section 721. Civil Code former section 158 provided that while spouses could transact with each other, those transactions were “subject... to the general rules which control the actions of persons occupying confidential relations with each other, as defined by the title on trusts.” We asked for, and received, supplemental briefing on these issues. We have already discussed how Haines and Mathews differ from the facts of this case. Benson, supra, 36 Cal.4th 1096 , concerned the rules governing the transmutation of community property pension benefits to separate property under Family Code section 852. The court said the husband had waived any issues under section 721 which was, in any event, inapplicable. (Benson, at pp. 1111-1112.) Delaney, supra, 111 Cal.App.4th 991 , concerned the invalidation of a deed by which the husband conveyed his separate property home to himself and his wife as joint tenants. The primary issue was who bore the burden of proof under section 721, with the court choosing to follow Haines . (Delaney, at p. 998.) On the merits, the court held the evidence supported a finding of undue influence because the husband had cognitive impairments and relied on the wife to handle their financial affairs. No promise to reconvey was at issue, and the undue influence therefore was of the coercive kind under Civil Code section 1575, subdivision 2.