Opinion

This appeal concerns the interpretation and administration of the Donald W. Callender Family Trust (Trust), executed by Donald W. Callender (Donald) as the settlor and original trustee in 2003. Defendant and appellant Cathleen Callender (Cathe), defendant and appellant Catherine T. Callender (Katy), and defendant Donald Lucky Callender (Lucky), the primary beneficiaries of the residue or residuary estate Trust (Beneficiaries), are at odds primarily about how the residue of the Trust is to be divided. The bulk of the assets which comprise the residuary estate are cash and other liquid assets, real property, and royalty agreements for licensing of the Marie Callender name. The Trust provided that upon Donald’s death, the residuary estate was to be divided into thirds and vested in each of the Beneficiaries.

Plaintiffs and respondents Douglas K. Ammerman (Ammerman) and Janet Feldmar (Feldmar) (collectively Trustees) succeeded Donald as the trustees upon his death in January 2009. When disputes arose among the Beneficiaries as to division of the residuary estate, they filed a petition for instructions.

After trial, the court ruled the Trust’s residuary estate should be divided based on what the parties referred to as the “changing fraction method.” The effect of this method is that when non-pro-rata payments and distributions were made, including most prominently estate tax payments by Cathe and Lucky only, the Beneficiaries’ fractional interests were to be revalued and changed from the original one-third allocation to each beneficiary.

The central issue on appeal is whether the court erred in ruling the changing fraction method applied to the residuary. Cathe argues, based on the terms of the Trust that initially divided the residuary estate into one-third interests for each of the three Beneficiaries, those percentage interests remained fixed and the changing fraction method does not apply. More specifically, she asserts the Beneficiaries’ percentage interests in the residuary estate did not change when the estate taxes were paid by Cathe and Lucky, or when other non-pro-rata payments or distributions were made.

Katy and Lucky argue to the contrary that the payment of the estate taxes immediately reduced Cathe’s and Lucky’s percentage interests in the Trust residuary to approximately 25 percent each, and simultaneously increased Katy’s percentage interest to approximately 50 percent. Katy and Lucky further contend those changed percentage interests in the Trust’s residuary estate must be applied to future payments and distributions, and must be recalculated every time non-pro-rata payments and distributions are made, using the changing fraction method.

This substantial difference in approach dramatically affects the ultimate division of the residuary assets of the Trust. And it is especially important as applied to the most valuable asset in the residue, the royalty payments resulting from licensing of the Marie Callender name, valued at $37 million on the federal estate tax return and perhaps worth substantially more now.

As a second issue, Cathe contends the court erred when it ruled she was responsible for a portion of the estate taxes on a piece of real property left to her outright, because the Trust stated she was not to be liable for any taxes.

We conclude Cathe is right on both of these issues. The court erred in ruling the changing fraction method applied to the Trust’s residuary estate, and that Cathe should be charged taxes on the gift of her property. Therefore we reverse the judgment.

Katy filed a cross-appeal, raising three issues. First, she disputes the revaluation of the assets in the residue after values were established for the federal estate tax return. Second, she challenges how the court applied an income tax overpayment. Finally, she claims it was error for the court to exclude more than $2.5 million attributable to her life estate in several pieces of residential real property from the value of the residuary estate. These issues are all based on application of the changing fraction method. Because we conclude the changing fraction method is not applicable and reverse the judgment, these issues are moot.

Cathe’s request for judicial notice regarding a related case filed by Katy is not relevant and we deny the request.

Additional facts are set out in the discussion.

FACTS AND, PROCEDURAL HISTORY

1. Terms of Trust and Distributions and Payments

Cathe is Donald’s daughter from his first marriage. Katy is Donald’s surviving widow. Lucky is Donald’s son with Katy.

On Donald’s death the assets of the Trust included cash, securities, real estate, and trademark license agreements (License Agreements) paying royalties (Royalties) for use of the name Marie Callender’s. The total value shown on the federal estate tax return was approximately $143 million. The most valuable assets were the License Agreements. From Donald’s death until the time of trial, $67 million in Royalties had been paid. The remaining Royalties were estimated to be worth between $64.8 million and $83 million.

The Trust provided for specific gifts, including real property in Corona del Mar (Goldenrod Property) to Cathe. (Art. IV, § B, par. 2b.) According to the Trust this gift “shall be free from any federal estate taxes, inheritance or succession taxes, generation skipping transfer taxes, or any other taxes payable upon or resulting from or by reason of the death of the settlor.” (Art. IV, § B, par. 2h.)

The Trust directed that “[o]n the death of the settlor, and after the [specific] distributions” are made, “the trustee shall divide the trust estate . . . into separate parts or shares. This division shall be made either by physical segregation of the assets or by assignment or transfer of undivided interests in the whole or any part of the trust property.” (Art. IV, § B, par. 3a.) The assets were to be divided, one-third each, to subtrusts for Cathe, Katy, and Lucky (Cathe’s Trust, Katy’s Trust & Lucky’s Trust, respectively; collectively Subtrusts).

The Trust allowed the Trustees to withhold payments to the Subtrusts during an initial holding period, although the Subtrusts continued to have the right to receive income from the assets. (Art. V, § F, par. 13.) The Trust further provided that “[w]henever the trustee is directed to make a distribution of trust assets or a division of trust assets into separate trusts or shares on the death of the settlor, the trustee shall have the discretion to defer such distribution or division until six months after the settlor’s death. However, the interest of the settlor’s spouse in any marital deduction trust or outright distribution qualifying for the marital deduction shall vest immediately on such settlor’s death.” (Art. V, § E, par. 17b.)

The Trustees did not either physically or on paper create the three Subtrusts and show assets going into those Subtrusts.

The Trust required that upon Donald’s death, the Trustees were to pay all estate taxes and apportion them to the Subtrusts. (Art. VIII, § A.) Katy’s Trust was not to “be reduced by any inheritance, estate or transfer taxes.” (Art. IV, § C, par. 2.) As a result when the Trustees paid the estate taxes they were to charge them to Cathe’s and Lucky’s trusts. (Art. VIII, § A.) The taxes were to be “collected from or charged to those persons before distributions from the [Trust] or any subtrust.” (Ibid.)

The Trust allowed the Trustees to make advances or loans to the Subtrusts. (Art. V, § E, par. 6.) Such an advance or loan was to bear interest at the current rate. (Ibid.)

In October 2009, the Trustees deposited $38.8 million for the federal estate taxes, and debited $19.4 million each of principal from Cathe’s Trust and Lucky’s Trust.

In January 2010 the Trust filed the federal estate tax return, showing the value of Katy’s Trust as $40,534,798 (47.714 percent), Cathe’s Trust as $22,479,541 (26.461 percent), and Lucky’s Trust as $21,939,541 (25.825 percent). After the IRS audit, there were two subsequent slight adjustments from a refund and then an additional payment.

In May 2012, the Trustees sent a letter to the Beneficiaries (May 2012 letter) proposing to distribute the residuary to the Subtrusts according to slightly revised percentages (46.795 percent, 27.042 percent, and 26.163 percent to Katy’s Trust, Cathe’s Trust, and Lucky’s Trust, respectively). They asked the Beneficiaries to agree to those percentages, which each of them did. The Trustees then distributed a total of $18,080,257 in principal to the Beneficiaries in those percentages. The Trustees had also distributed tens of millions of dollars in unequal proportions to the Beneficiaries when they requested or needed a distribution. By December 2012, distributions totaled $39,760,510.91 from Katy’s Trust, $21,041,956.52 from Cathe’s Trust, and $18,170,251.86 from Lucky’s Trust. Payments on account of the estate taxes are not included in these amounts.

Jack Barcal was the attorney for administration of the Trust. He had been Donald’s lawyer and drafted the Trust. Barcal is a certified public accountant (CPA) and certified specialist in tax and in estate planning, probate, and trust. He practices in that field and is also a full-time professor at USC (University of Southern California) teaching tax theory, estate and gift taxation, advanced estate planning, and family wealth preservation. He regularly speaks at estate and gift tax seminars and conferences and publishes annually. He described the Trust as large.

In October 2012, after speaking with representatives of Cathe and Katy regarding distribution of the Trust assets, Barcal sent a letter to the Trustees stating he had “decided to start from scratch and review the [Trust] re Don’s intentions.” He explained Donald “[c]learly” wanted the assets divided one-third each to the Beneficiaries and that Cathe and Lucky would pay the estate taxes. He stated he had examined the allocations and distributions to date and concluded “the allocation percentages should be 1/3 each because there were sufficient liquid assets that should have been distributed to each beneficiary,” $19 million each to the IRS on behalf of Cathe and Lucky and $19 million to Katy. All of the remaining assets should have been divided pro rata, one-third each “as provided in the [Tjrust.” “The prior allocation percentages were based on liquid assets not being sufficient to pay the estate taxes, which was not the case.”

There has been no final accounting.

2. Petition for Instructions

The Trustees subsequently filed a petition for instructions as to how the Trust’s residuary estate should be distributed. They alleged the carrying value of the then remaining residuary assets was approximately $75 million and the fair market value was significantly more. They also alleged Barcal initially advised the Trustees to deduct the estate taxes from Cathe’s and Lucky’s shares. That had the effect of reducing their shares of the Trust’s residuary estate.

According to the petition, Barcal subsequently told them that initial allocation was not correct. Rather, the Trustees should divide all the residuary assets into thirds, and then allocate one-half of the estate taxes each to Cathe’s and Lucky’s trusts. This would ensure the Beneficiaries would each be receiving one-third of the residuary assets. This reduced the amount of the residuary assets payable to Katy.

In her response to the petition, Cathe asked the court to instruct the Trustees to distribute the residuary assets or estate equally in thirds to the Beneficiaries according to the Trust’s terms, after making adjustments for tax payments.

Katy’s response asked the court to instruct the Trustees to distribute the residuary in the following percentages: Katy, 46.795 percent, Cathe, 27.042 percent, and Lucky, 26.163 percent. Lucky prayed for the same relief.

3. Trial

There was a multiday trial on the petition. Ammerman, one of the Trustees, testified he had been a CPA for close to 30 years, ultimately managing the Orange County office of Peat, Marwick, Mitchell, before retiring. He is not a professional fiduciary. He met Donald in the mid-1970’s and in the mid-1980’s Donald became a client.

Ammerman testified Donald knew the amount of distributions to Cathe and Lucky would be reduced by the estate taxes but never told him Katy’s share would be more than one-third of the assets. Ammerman also testified he had never heard of the changing fraction method before Katy raised it during the litigation and had not administered the Trust using that method. Instead, he had used the same percentages as shown on the estate tax return. He testified he relied on the advice of the lawyers and accountants in making distributions.

Feldmar, the other Trustee, was also a CPA. She began her career working with a large accounting firm and subsequently started her own firm. She met Donald in the mid-1980’s and later prepared his personal tax returns. She is not a professional fiduciary and, although she is named as a trustee in several trusts, she has never served in that capacity except for the Trust. She was named cotrustee at about the time the Trust was executed.

She too testified she had never heard of the changing fraction method before this action was filed. She reiterated Donald had told her his intent was to distribute the Trust’s residuary estate equally in one-third shares to the Beneficiaries. She did not recall ever discussing with him an uneven allocation of the residuary assets.

Likewise, Barcal, the attorney who drafted the Trust, testified he had never heard of the changing fraction method until this litigation. He stated: “You never get into any of this changing fraction stuff, which nobody in the world uses. Never seen anybody use it. And especially when you don’t have the intention of the testator.”

Barcal testified he told Donald that although the assets would vest in the Beneficiaries’ trusts upon Donald’s death, the Trustees did not have to divide them until the Trust administration concluded. He also testified the Trust allowed for unequal distributions and such distributions did not change the one-third vested interests. The distributions would be accounted for in the final accounting. The same was true for the tax payments paid by Cathe’s Trust and Lucky’s Trust.

Barcal believed that was Donald’s intent and was what the Trust stated. He thought it was “outrageous” Katy would receive almost 48 percent of the Royalties because it violated the Trustees’ duty of impartiality.

Katy’s expert was Robert Goodrich. He had been a CPA for 40 years, specializing in estate and trust administration, preparing over 1,000 fiduciary tax returns and hundreds of accountings, estate tax returns, and allocations of funding for subtrusts. During all that time, before this case he had never applied the changing fraction method. This was because there had been no non-pro-rata principal distributions during administration. Although he relied on Probate Code sections 16340 and 16341 (all further statutory references are to this code unless otherwise stated) as the basis for applying the changing fraction method, he could not point to any portion of those sections showing they applied to distribution of principal.

Jan Goren was Cathe’s accounting expert. He had been a CPA for 35 years and was also certified in financial forensics. As an expert witness, he testified in cases involving, among other things, accounting issues and analysis, and tracing of assets. He was a trustee of three trusts with assets and 10 or so others not funded. He has also served as a court-appointed referee, primarily on cases where there are numerous transactions and between 50 to 60 accounting issues.

Goren testified as to the “hazard[s]” of using the changing fraction method, showing Goodrich’s application lacked “computational integrity.” For example, Goodrich used a cost basis instead of fair market value of assets and did not take into account certain distributions. Goren reconciled and adjusted the changing fraction method of accounting to show the difference if distributions were made on a one-third basis.

Goren testified payment of the estate taxes should not have affected the one-third formula set out in the Trust. There were enough liquid assets to pay the estate taxes from Lucky’s Trust and Cathe’s Trust, respectively, and to make a payment in the same amount to Katy.

Goren also testified the Trust contained a provision allowing the Trustees to make an advance or loan to a beneficiary, subject to interest, making it unnecessary to change the one-third division of assets. He prepared schedules showing how this would be done.

Richard Heaton, Lucky’s Trustee ad litem, was a CPA and had been an attorney for almost 40 years, specializing in trusts and estates. He has advised trustees and prepared fiduciary accountings.

Heaton testified a trustee may elect to use the changing fraction method depending on the circumstances. He believed the changing fraction method was more equitable than using a set fractional method in this case because “the risks of depreciation and appreciation are unequally shared by various parties.”

4. Statement of Decision

After trial, the court issued a statement of decision. It ruled the Trustees should use the changing fraction method, not the “fixed fraction with interest” compensation method. The need for choosing between these two methods arises “only when the trustees make non[-]pro[-]rata charges to principal,” as in this case, and thus the beneficiaries’ fractional interests must be determined.

Pursuant to the changing fraction method, all “non[-]pro[-]rata charges against principal result in changes to the fractional interests of the residuary beneficiaries, with later income, principal receipts and gain (or loss), whether realized or unrealized, accruing to the residuary beneficiaries pro rata based on the new fractional interests.”

Under the fixed fraction with interest method, on the other hand, “non[-] pro[-]rata charges to principal have no effect upon the fractional interests of the beneficiaries in future receipts and gain (or loss), whether realized or unrealized, so that these accrue to the respective beneficiaries pro rata based on their original fractional interests.”

To arrive at the conclusion the changing fraction method applied, the court started with the premise the settlor’s intention as stated in the Trust “ ‘controls the legal effect of the dispositions made in the instrument.’ ” (Quoting § 21102, subd. (a).) To determine that intent, the court first looked to the Trust’s terms. (Safai v. Safai (2008) 164 Cal.App.4th 233, 244 [ 78 Cal.Rptr.3d 759 ].)

The court found article IV, section B, paragraphs 3 through 3.2 clearly stated the estate’s residue was to be divided into the three separate Subtrusts, and under paragraph 3a the interests vested on Donald’s death. The vesting was “subject to the prior distribution of specific bequests and the prior payment of certain expenses” as set out in the Trust.

The court decided the Trust provided Katy’s Trust was not to be subject to any taxes due on account of Donald’s death, leaving taxes to be charged against Cathe’s and Lucky’s trusts.

The court also found article VIII required payment of taxes and other items to be made before the Subtrusts were funded. It interpreted this to mean payment of estate taxes would have the effect of reducing Cathe’s and Lucky’s shares in the residue and concluded Donald would have known and intended this would change the equal thirds ratio.

Nevertheless the court stated it was not clear from the Trust whether the settlor “contemplated the effect of the one-time change of fractional interests” or “contemplated that future non[-]pro[-]rata charges against principal might raise the same issue, or what his intentions” might have been. The court could not find a trust provision that answered these questions. Therefore, the court looked to “additional guidance outside the language of the Trust.”

It first reviewed sections 16340 and 16341. Section 16340, subdivision (d) states that, after some preliminary steps, the trustee must distribute net income as set out in section 16341. Section 16341 states “[e]ach beneficiary described in subdivision (d) of Section 16340 is entitled to receive a portion of the net income equal to the beneficiary’s fractional interest in undistributed principal assets, using values as of the distribution dates and without reducing the values by any unpaid principal obligations.” The court stated this meant that if there was no contrary term in the Trust, the residuary Beneficiaries’ fractional interests in income are tied to their fractional interests in principal. The court decided these sections provided for use of the changing fraction method for income.

The court recognized that an important component of the Trust is principal receipts realized from the Royalties. It agreed with Cathe’s counsel that sections 16340 and 16341 did not apply to principal receipts “explicitly.”

The court held, however, section 16341, subdivision (a) “plainly recognize[d] both the possibility that fractional interests in principal can change, and the need to allocate receipts to the beneficiaries on the basis of their changed fractional interests in undistributed principal assets, rather than any original or prior interests therein.” Absent anything to the contrary, the court could not find any reason to “give [principal receipts] special treatment.”

The court discounted Cathe’s argument the Trust preempted application of sections 16340 and 16341. The court explained it applied the sections only because the Trust did not answer the question of “the effect of non[-]pro[-] rata charges against principal.”

In concluding the changing fraction method applied, the court also relied, without comment, on four cases from other states, three law review articles, and a section of the Internal Revenue Code and a related Treasury regulation.

In addition, the court found using the changing fraction method was more equitable, in the absence of Donald’s clear intention to the contrary. If one beneficiary has a greater stake than another, that party should be allocated the gain and bear the burden of any loss in the value of the assets. This is true even though the Beneficiaries did not realize at the time that taking a non-pro-rata distribution would affect their fractional interests. Likewise, it would seem inequitable if a beneficiary expected the changing fraction method to be applied and it was not. The court’s ruling was not conditioned on any beneficiary’s mental state. The court thought it significant the Trustees had initially applied “a rudimentary form” of the changing fraction method as to the effect of the estate taxes and no beneficiary had objected. This was reflected in the May 2012 letter, which the Beneficiaries each signed.

The court found the May 2012 letter was not “legally binding,” was not supported by consideration for the Beneficiaries’ signatures, did not constitute an amendment to the Trust, and was not a consent to the method of calculation or a waiver of the right to object to it. Even so, it did show the equity in applying the changing fraction method and was “meaningful evidence” because the parties did not object to the calculations at the time.

The court also relied on the expert testimony to support use of the changing fraction method. Goodrich favored it. Although Goren did not, he approved of its use in other situations, including as to appreciating stocks. The court found little difference between appreciating stocks and the Royalties.

The statement of decision provided that, despite the Trustees having the authority to make loans or pay advances, it would be a “fiction” to find payment of the estate taxes from Cathe’s and Lucky’s trusts was an advance and a “fabrication]” to find a retroactive equalizing payment to Katy when the estate taxes were charged to Cathe and Lucky. The court also found it “counterintuitive and illogical” and contrary to the terms of the Trust to wait until the final accounting to charge estate taxes to Cathe and Lucky.

The court was not concerned with the arguable difficulty of valuing the Royalties over the course of administering the Trust.

The court set out a formula for how the Trustees were to apply the changing fraction method. The first valuation would be as of the date of Donald’s death and then to be recalculated quarterly and also when there was a material principal distribution. Appraisals were also to be done at those same times. The fractional interest of each residuary beneficiary should be changed with each recalculation.

The changing fraction method should then be applied as follows: using the above formula, the Trustees should calculate net principal receipts allocable to each beneficiary and add it to the beneficiary’s previous share. Next, any principal charges allocable to each beneficiary, including principal distributions and non-pro-rata principal distributions, should be determined and subtracted from a share as just calculated.

This establishes the adjusted value of each beneficiary’s share. When compared to each other, these determine the changed fractional interests. Those new interests are the starting point for the next period.

Finally, after arriving at the new fractional interests, the Trustees are to add any unrealized appreciation in the Trust assets and subtract unrealized loss from the value of the undistributed Trust assets for each beneficiary.

As to the Goldenrod Property, the court ruled the Trust provision that Cathe owes no estate taxes meant only she was not to be charged with the full portion of the estate tax. It found Donald’s intent was that her estate tax payment should be no more and no less than if the property had been given to a third party. The court found no evidence Donald wanted Cathe to have a greater share in the residuary than Lucky. It understood Cathe would bear half of the burden of the estate taxes for the Goldenrod Property.

DISCUSSION

1. Standard of Review

The parties dispute the standard of review to be applied. Cathe argues for de novo review because the case turns on interpretation of the Trust where the court found no ambiguity and did not weigh any conflicting extrinsic evidence. Lucky agrees “many of the key considerations,” though unnamed, are subject to de novo review, but argues the court relied on some extrinsic evidence. As to that, he claims, the standard is substantial evidence. He further asserts that in deciding whether the court correctly applied the changing fraction method, the standard is abuse of discretion. Katy maintains that because the case arises out of the petition for instructions, the proper standard is abuse of discretion.

This case turns on interpretation of the Trust, which is a question of law we review de novo, unless there is a conflict in extrinsic evidence. (Siegel v. Fife (2015) 234 Cal.App.4th 988, 995-996 [ 184 Cal.Rptr.3d 531 ].) “ ‘Extrinsic evidence is “admissible to interpret the instrument, but not to give it a meaning to which it is not reasonably susceptible” [citations], and it is the instrument itself that must be given effect. [Citations.] It is therefore solely a judicial function to interpret a written instrument unless the interpretation turns upon the credibility of extrinsic evidence.’ [Citations.]” (Id. at p. 996.)

Here testimony and documentary evidence were presented at trial. However, the statement of decision does not identify any conflicting extrinsic evidence as to the meaning of the Trust. Nor have the parties pointed to conflicting extrinsic evidence as to the interpretation of the Trust. Katy affirmatively states the Trust provisions are not in dispute.

There was a conflict in expert testimony about whether the changing fraction method should be used, but this was only about accounting and distribution theories, not whether the changing fraction method was provided for in the Trust. Although the court did rely on some expert testimony, the question presented is not whether that testimony was sufficient to support the judgment, as Lucky contends.

Further, contrary to Katy’s and Lucky’s arguments, the primary issue is not whether the court properly exercised its discretion in instructing the Trustees. Those instructions were the direct result of the court’s interpretation of the Trust. Rather, as stated above, the central issue is whether the changing fraction method should be applied to the residuary assets.

2. Principles of Trust Interpretation

In interpreting the Trust, our prime consideration is Donald’s intent. “The paramount rule in construing such an instrument is to determine intent from the instrument itself and in accordance with applicable law. [Citations.]” (Brown v. Labow (2007) 157 Cal.App.4th 795, 812 [ 69 Cal.Rptr.3d 417 ]; see § 21102, subd. (a) [“The intention of the transferor as expressed in the instrument controls the legal effect of the dispositions made in the instrument.”].) “ ‘ “In construing a trust instrument, the intent of the trustor prevails and it must be ascertained from the whole of the trust instrument, not just separate parts of it.” ’ ” (Estate of Cairns (2010) 188 Cal.App.4th 937, 944 [ 115 Cal.Rptr.3d 735 ].) Where the evidence is undisputed but conflicting inferences may be drawn, we “ ‘independently draw inferences.’ ” (Siegel v. Fife, supra, 234 Cal.App.4th at p. 996 .)

“ ‘The interpretation of a written instrument, even though it involves what might properly be called questions of fact [citation], is essentially a judicial function to be exercised according to the generally accepted canons of interpretation so that the purposes of the instrument may be given effect. [Citations.] Extrinsic evidence is “admissible to interpret the instrument, but not to give it a meaning to which it is not reasonably susceptible” [citations], and it is the instrument itself that must be given effect. [Citations.] It is therefore solely a judicial function to interpret a written instrument unless the interpretation turns upon the credibility of extrinsic evidence. . . . “An appellate court is not bound by a construction of the contract based solely upon the terms of the written instrument without the aid of evidence [citations], where there is no conflict in the evidence [citations], or a determination has been made upon incompetent evidence [citation].” ’ ” (Gardenhire v. Superior Court (2005) 127 Cal.App.4th 882, 888 [ 26 Cal.Rptr.3d 143 ], italics added.)

“[I]n ascertaining the intention of the trustor the court is not limited to determining what is meant by any particular phrase but may also consider the necessary implication arising from the language of the instrument as a whole.” (Brock v. Hall (1949) 33 Cal.2d 885, 890 [ 206 P.2d 360 ]; see § 21120 [“The words of an instrument are to receive an interpretation that will give every expression some effect, rather than one that will render any of the expressions inoperative.”]; see also §21121 [“All parts of an instrument are to be construed in relation to each other and so as, if possible, to form a consistent whole. If the meaning of any part of an instrument is ambiguous or doubtful, it may be explained by any reference to or recital of that part in another part of the instrument.”].)

In addition, we need to “ ‘ “consider the circumstances under which the document was made so that the court may be placed in the position of the . . . trustor whose language it is interpreting, in order to determine whether the terms of the document are clear and definite, or ambiguous in some respect.” ’ [Citation.]” (Estate of Powell (2000) 83 Cal.App.4th 1434, 1440 [ 100 Cal.Rptr.2d 501 ].)

3. The Changing Fraction Method Does Not Apply.

a. Overview

As we will explain below, we conclude there is simply no basis for applying the changing fraction method to the Trust. It is not found in the Trust provisions. There is no evidence to show this was Donald’s intent. In fact, the Trust and the circumstances surrounding its execution show just the opposite. The Trust plainly stated Donald wanted the residue divided one-third each among the Beneficiaries. Although the estate taxes were allocated to Cathe and Lucky, the Trust did not provide this allocation would alter the one-third division or the vested interests of the Beneficiaries in their one-third shares of the residuary estate. The testimony of Barcal, as the drafter of the Trust, is consistent with the Trust language and is highly probative.

No California case and nothing in the Probate Code applies or even discusses the changing fraction method. Application of the changing fraction method here is importing into California law a principle that has never officially been sanctioned. There is no justification for doing so in this case.

Nor does equity justify application of the changing fraction method. The basis of the court’s finding was its assumption the fractional interests of the parties were not equal. But that is not supported by the language of the Trust or Donald’s intent.

If the Trust is interpreted according to its terms, and honoring Donald’s intent, there is no need and no place for the changing fraction method. The court’s application of the changing fraction method started with a false premise that after payment of the estate taxes, the Beneficiaries’ fractional interests were no longer one-third each and there was nothing in the Trust to the contrary. All of the court’s justification for its decision flowed from that false premise. That is why we reverse.

b. The Trust Language Does Not Support the Changing Fraction Method.

Pursuant to the Trust, on Donald’s death and after certain enumerated specific distributions, the Trust required the Trustees to “divide ... the trust estate” into the three separate Subtrusts. (Art. IV, § B, pars. 3, 3.1, italics added.) The court found such to be the case and that “the interests of the three beneficiaries vested upon the death of the settlor.” There is no dispute as to these findings.

Having made these findings, however, the court noted the residuary assets were never divided into three Subtrusts, and analyzed the residue as one undifferentiated pool of assets. It referred to “the stake that each party has in the underlying asset pool,” finding it was “natural” a beneficiary with a “greater stake than the others” “should benefit from income, realized gains, appreciation, and so forth to a greater extent” and also “suffer” the effect of “adverse events.” The court’s findings as to the vesting and how the residuary estate should be divided misinterpreted the Trust. It was using this mistaken conceptual treatment of the residuary estate, i.e., the assets were not divided into the Subtrusts, at least before payment of estate taxes, that the court found the changing fraction method applied.

The reason this matters is because the Trust dictated that Cathe and Lucky pay the estate taxes with no charge to Katy. (Art. IV, § C, par. 2; Art. VIII, § A.) The court found the Trust required the estate taxes to be paid before funding of the Subtrusts, and this requirement suggested use of the changing fraction method.

It concluded Donald “must have understood that this would result in a reduction of the total principal assets of Cathe[’s] and Lucky’s [S]ubtrusts, so that the actual interests of the residuary beneficiaries after taxes, even before the actual funding of the [S]ubtrusts, would change from the 1/3-1/3-1/3 ratio set forth in Article IV, Section B, Paragraphs 3.1.” According to the court, this showed Donald’s intent there be different proportionate amounts and “suggested] the possibility that he intended a changing fraction to apply to the Trust.” (Italics omitted.) These findings conflict with the terms of the Trust and are at the core of the underlying dispute.

The Trust provided that on Donald’s death the Trustees were to pay estate taxes, which were to be charged to Cathe and Lucky “before distributions from the [Trust] or any subtrust.” (Art. VIII, § A, italics added.) This language presumes there would be assets in the Subtrusts from which distributions could be made. This section plainly does not state the Subtrusts are to be funded after payment of estate taxes, as the court held.

Dividing the assets into the Subtrusts before payment of the estate taxes, as the Trust mandated, eliminated any necessity of using a changing fraction method. But the court proceeded on the contrary assumption the $38 million in estate taxes was to be paid from one large pool of undivided assets. Thereafter, upon funding the Subtrusts, Katy would be entitled to an additional $19 million worth of assets, leading to a change in the proportional interests.

By funding the Subtrusts first, $19 million each for the estate taxes would be charged to Cathe and Lucky, and Katy’s $19 million would remain in her Subtrust until it was distributed to her. Receipts and gains and losses would then be credited to or charged against the Subtrusts in the one-third ratios. Thus, Cathe’s and Lucky’s payments of the estate taxes would reduce the amount of their interests but not affect their fully vested one-third interests in the other nonliquid interests, including the Royalties to be paid under the License Agreements.

The residuary consisted of liquid assets, valued at approximately $66 million, real estate, valued at approximately $22.5 million, miscellaneous assets, valued at just over $5 million, and the License Agreements, valued for estate tax purposes at $37 million but also valued at between $51.6 million and $78.5 million.

Cathe contends that on Donald’s death when the Subtrusts were created and the assets vested, each of the four categories of assets, i.e., liquid assets, real estate, the License Agreements, and miscellaneous assets, was to be divided, one-third to each of the Beneficiaries. This would include $22 million each of liquid assets, enough to pay the estates taxes without requiring the sale of any assets.

Lucky argues that the provision calling for division of the residuary assets into thirds (art. IV, § B, pars. 3, 3.1) “contains no specific instructions as to how to divide any particular asset” and that Cathe is demanding “an arbitrary division of residuary assets” between liquid and nonliquid so each Subtrust holds an equal amount of each and further that the estate taxes be charged against the liquid assets. He contends the paragraph “must be read as giving to each [Sjubtrust a 1/3 vested interest in the total, undifferentiated residue of the Callender Trust.”

Lucky is mistaken. The paragraph plainly stated “[e]ach part or share shall be administered as a separate and distinct trust.” (Art. V, § B, par. 3a.) Moreover, other Trust provisions support an interpretation there be three separate, funded Subtrusts. For example, the Trust gave Katy the right to withdraw all of her share of the assets at any time. (Art. IV, § C, par. 6c.) Further, the Trustees have the power to “invest the funds of each separate trust estate created by this instrument either in property owned solely by the respective trust estates or in property owned jointly by all of the trust estates.” (Art. V, § E, par. 15.) Not only that, the Trustees’ “[Retention of a beneficiary’s share [pending payment of estate taxes] shall not affect his or her right to income therefrom, if any, or his or her power of disposition over such property.” (Art. V, § E, par. 13.) These provisions would not make sense if the residuary assets remained in one large pool as opposed to being divided into three separate Subtrusts.

Lucky also argues the provision allowing for non-pro-rata distributions (Art. V, § E, par. 17a) defeats Cathe’s argument the Beneficiaries each had a one-third interest in each asset. He concedes dividing the liquid assets one-third each and the nonliquid assets the same way might make sense, but maintains the Trust did not provide for that.

Admittedly, that paragraph gave the Trustees the discretion to make non-pro-rata divisions or distributions. But they may also make non-pro-rata divisions between the Subtrusts “as long as the respective assets allocated to separate trusts or shares, or distributed to such beneficiaries, have equivalent or proportionate fair market value.” (Art. V, § E, par. 17a.) If the Beneficiaries’ interests were supposed to be held in an undifferentiated pool of assets, there would be no need for this provision.

In exercising their discretion, the Trustees must still uphold their duties as fiduciaries and distribute the assets fairly. (§ 16003 [where trust has more than one beneficiary, trustee must “deal impartially with them and shall act impartially in investing and managing the trust property, taking into account any differing interests of the beneficiaries”]; Penny v. Wilson (2004) 123 Cal.App.4th 596, 603-604 [ 20 Cal.Rptr.3d 212 ] [even with absolute discretion trustee must act in good faith and, if required by trust, ensure distribution is equal].) Nothing in the Trust supports the Trustees’ use of the changing fraction method in the exercise of their discretion.

Again, in interpreting the language of the Trust, our charge is to determine the intent of the trustor. Nothing in the four corners of the Trust supports the changing fraction method, and the testimony of Barcal, the attorney who drafted the Trust, unequivocally supports the conclusion Donald did not intend to use it.

Barcal testified, “Don really wanted a third to each.” Donald did not intend the one-third divisions would change upon payment of the taxes. “Just the opposite. That he really, in an ideal world, would prefer one-third each after the taxes. And no one could tell exactly what they were, but it was going to be a third each.” Donald told him, “okay, let [Katy] have her third and let. . . [Cathe and Lucky] just pay the taxes.”

Further, Barcal stated, there were enough liquid assets in the Trust such that taxes could be paid from Cathe’s and Lucky’s shares “and you don’t get into any of this as long as you make the principal allocations after the date of death, which is what happens with vesting in the subtrusts. [¶] You never get into any of this changing fraction stuff, which nobody in the world uses. Never seen anybody use it. And especially when you don’t have the intention of the testator.” Barcal confirmed he had never heard of the changing fraction method to be used in distributing principal when a trustor’s intent was stated in the trust.

Barcal explained vesting of the Subtrusts occurred upon Donald’s death, one-third each, in every asset. In discussions with Donald about creating the Trust he had explained the one-third division and the vesting at the time of death. He also testified unequal distributions were allowed by the Trust and they did not affect or change the one-third divisions. “And then if you want, the Trustees can make adjustments later on along the way.” They would be reconciled in the final accounting.

Regarding the estate tax payments by Cathe and Lucky, Barcal agreed that, although the taxes were paid in October 2009, the Subtrusts were not charged until the final accounting. “You do that at the end. . . . [Y]ou make adjustments. You do that in every big estate. ... It’s just normal. That’s why you have that closing period provision in the Trust. It’s not intended to interfere with the vesting in any way and it can’t.”

Similarly, Feldmar, one of the Trustees, testified Donald had told her he intended the residue would be divided into equal thirds, and she had no recollection he had ever spoken of an unequal allocation. Further, she knew nothing of the changing fraction method until this litigation.

Ammerman likewise testified Donald knew the amount of final distributions to Cathe and Lucky would be lower than those to Katy because of tax payments due by the children. But Donald never said Katy’s share would be greater than one-third of the assets. Moreover, Ammerman had never heard of the changing fraction method and had not relied on it to administer the Trust.

Thus, it could not have been Donald’s intent to use the changing fraction method. Barcal, who drafted the Trust, had never heard of it, so it could not have been included, and there is no evidence Donald knew of or intended it be used in the Trust. In fact, the first time the term was used among the parties was in a letter sent by Katy’s counsel after this action was filed.

Katy points to her testimony that the day Donald signed the restatement of the Trust, he told her he chose a “spousal or marital deduction,” which meant that after Cathe and Lucky paid the estate taxes, the residue of the Trust would be “50, 25, 25.” She claims that on another occasion he told her she would be inheriting enough money, i.e., “roughly 50 percent of this estate, [that she] should never have to fly an airplane again.” But, in their conversations witih Katy, neither Ammerman nor Feldmar remembered ever discussing Donald’s alleged statements that the ultimate trust split would be 50-25-25.

Katy’s testimony, which the statement of decision does not mention, does not show it was Donald’s intent to use the changing fraction method. First, Donald’s purported hearsay statements to her do not refer to the changing fraction method. In addition, in light of the Trust provision of a one-third division among the Beneficiaries, in order to guarantee a 50-25-25 split, Donald would have had to have known, among other things, the date of his death, the value of the estate on that date, and the estate tax provisions in effect at that time. Barcal confirmed “it was impossible, considering that Donald was well alive, for anybody to know on [the date the Trust was restated], what estate tax rate would be paid at his death.” Donald and Barcal discussed that fact. Katy and Lucky are merely trying to imply the changing fraction method into that testimony.

Lucky challenges the testimony that none of these witnesses had heard of the changing fraction method. He contends this is merely a “semantic shell game,” that does not dispel the fact the Trustees initially administered the Trust in accordance with that principle, allocating principal in a 50-25-25 ratio. But again the question is whether Donald intended the Trust to use the changing fraction method, not whether the Trustees administered the Trust in that fashion. And the fact Barcal never discussed the changing fraction method with Donald and testified Donald never intended to use it, buttresses our conclusion it was not provided for in the Trust.

Katy and Lucky also dispute that the Trust required the immediate division of the assets into the Subtrusts. They rely especially on two provisions. First, the Trust stated when the Trustees are “directed to make a distribution of trust assets or a division of trust assets into separate trusts or shares on the death of the settlor,” the Trustees may defer distribution or division for a period of six months. (Art. V, § E, par. 17b.)

They also point to a provision which allowed the Trustees to “hold, administer and manage the [Trust] during a closing period which shall continue until” the amount of taxes has been determined, and the Trustees “may withhold from distribution . . .. all or any part of the property . . . subject to conflicting claims, to tax deficiencies, or to liabilities . . . .” (Art. V, § F, par. 13.)

Katy claims if the residuary assets were to be divided immediately into the three Subtrusts there would be no need for an administrative trust during the initial period following death.

Perhaps Katy’s claim would make sense if these were the only provisions dealing with this subject. But they are not. And this interpretation cannot be reconciled with the utter lack of any mention of the changing fraction method in the Trust or in the extrinsic evidence of Donald’s intent. Rather, we are persuaded that reading all the Trust provisions together, as we must, it makes sense to have an administrative period allowing the Trustees to marshal the Trust’s assets and withhold distribution during the closing period. But the Tmstees must still treat the residuary assets as if divided as of the date of death.

Katy and Lucky also rely on a provision allowing the Trustees to “make non-pro[-]rata division between trusts or shares and non-pro[-]rata distributions to [the] beneficiaries.” (Art. V, § E, par. 17a.) They argue this undermines the claim the Trust required the Subtmsts to be funded immediately. If that were the case, they contend, there would be no need to allow for non-pro-rata distributions.

Again, we are not persuaded. As Cathe explains, the provision gives the Trustees flexibility in allocating assets to each Subtrust, including the authority to exchange assets between the Subtrusts once a non-pro-rata distribution is made. Article V, section E, paragraph 17a provides non-pro-rata divisions and distributions may be made “as long as the respective assets allocated to separate trusts ... or distributed to such beneficiaries, have equivalent or proportionate fair market value.”

Katy cites IRS Revenue Ruling No. 69-486 (1969-2 CB 159) to support her claim the Beneficiaries own a fractional share of the undifferentiated pool of assets rather than a share in each asset. This ruling deals with the tax treatment of non-pro-rata distributions based on whether a trustee has the authority pursuant to the trust to make such distributions. (Ibid.) Katy argues that because the Trust gives the Trustees the authority to make non-pro-rata distributions, the Trust therefore provides the Beneficiaries do not own an interest in each asset and taxes would be triggered upon such distributions.

This argument does not persuade. Without commenting on the revenue ruling, the Trust plainly does not support Katy’s conclusion. Thus the tax effects, if any, of distributions do not affect Donald’s intent or the meaning of the Trust.

Cathe relies on a provision allowing the Trustees to “advance or loan funds” to a beneficiary, with an interest charge until paid. (Art. V, § E, par. 6.) She argues charges to pay the estate taxes were such advances, pointing to the provision taxes are to be “collected from or charged to [Cathe and Lucky] before distributions from the . . . Trust or any subtrust.” (Art. VIII, § A.)

Cathe contends that had the Trust intended use of the changing fraction method, there would have been no need for this provision. Thus, application of the changing fraction method in denigration of this section would violate the rules of trust interpretation that require us to give meaning to every provision in the Trust, in conjunction with all others, and not render one superfluous. (§§ 21120, 21121; Estate of Gregg (1976) 54 Cal.App.3d 882, 894 [ 127 Cal.Rptr. 64 ] [interpretation may not render trust provision surplusage].) We agree. This is proper accounting

The court found it would be a fiction to characterize Cathe’s and Lucky’s payments of the taxes as advances or loans to avoid using the changing fraction method. Katy and Lucky agree. We disagree.

Lucky further argues the Trustees treated payment of estate taxes as having been “collected from” the Subtrusts and not as advances. The residuary interests on the estate tax return showed Cathe’s and Lucky’s interests lower as a result of paying those taxes. Ammerman testified the Trustees did not “consider[]” treating the tax payments as loans to Cathe’s and Lucky’s trusts. And, Lucky goes on, there were no promissory notes to evidence a loan.

It is true, the Trust did not require the charges for tax payments be treated as advances or loans. But it did not prohibit it either. That the Trustees did not consider these payments loans or advances does not mean they cannot be characterized that way or that the Trustees were correct. Additionally, the fact no note was executed is not material. The Trust provides interest will be charged, and it can be accounted for in the final accounting. Non-pro-rata distributions can be handled that way as well.

Barcal testified that at the final accounting, adjustments are made, mistakes are corrected. This is the way large estates are handled. “That’s why you have [a] closing period provision in the Trust. It’s not intended to interfere with the vesting in any way and it can’t.”

Lucky and Katy dispute this interpretation and argue nothing in the Trust provides for truing up. Katy claims it is contrary to Donald’s intent and the Trust’s language, as well as Barcal’s and the Trustees’ actions for four years.

In our reading of the Trust we do not find specific support for truing up. But even assuming truing up is not specifically set out in the Trust, nothing in the Trust prohibits its use. And the absence of such a specific provision in no way supports use of the changing fraction method.

Moreover, accounting for payments from the residue to or on behalf of the Beneficiaries in this fashion is consistent with Donald’s intent those assets be divided and paid out one-third each to the Beneficiaries. The provision allowing for loans and advances is a legitimate method to effect that intent.

Katy argues there is “overwhelming” evidence that in administering the Trust, the Trustees “applied what they thought was Donald’s intent,” i.e., use of the changing fraction method. She points to several e-mails and letters they sent or received and schedules they prepared showing the roughly 50-25-25 division. They also testified they understood this was how the residue was to be divided.

Ammerman explained he thought the Beneficiaries owned the assets in roughly a 50-25-25 percent division. Feldmar testified she believed after death taxes were charged to Cathe’s and Lucky’s trusts, the fractional shares would change. She also understood the Beneficiaries’ rights to principal would be the same as their rights to income based on the final calculation of their fractional shares.

But that does not mean the Trustees’ interpretation of the Trust was correct or complied with the Trust’s provisions. We must construe the Trust to effect Donald’s intent as expressed in it. (§ 21102.) Moreover, several of the e-mails and letters do not prove the Trustees were relying, even unknowingly, on the changing fraction method. Some showed they were not at all certain of how to distribute the residue. In addition, as Cathe points out, the Trustees made several distributions of principal totaling millions of dollars, not at all in a 50-25-25 percent ratio.

Katy and Lucky also emphasize Barcal sent correspondence supporting the Trustees’ past distributions. But Barcal testified he told the Trustees their actions to date could be dealt with in the final accounting. For example as to the May 2012 letter, he told the Trustees “they were still okay, as long as [they] distributed the $19 million of cash [to Katy].” “[T]hen the assets would be distributed a third, a third, a third.”

Further, even if Barcal was initially mistaken, including in some of his communications with the Trustees and the parties, at some point upon his review of the Trust administration, he realized the 50-25-25 percentages were wrong. He noted, for example, they had been based on the incorrect belief there had not been sufficient liquid assets to pay the estate taxes. But the Trustees’ payments based on the incorrect percentages could be trued up in the final accounting.

In sum, nothing in the Trust supports use of the changing fraction method. On the date of Donald’s death each beneficiary had a vested one-third interest in all of the Trust residuary, including future receipts, regardless of whether they are characterized as principal, interest, or income. So the 50-25-25 and changing fraction splits of future receipts is inconsistent with the one-third each division expressly required by the Trust.

c. California Law Does Not Support Application of the Changing Fraction Method.

The Trust provided it is governed by the laws of California. (Art. V, § F, par. 4; see § 21103 [“The meaning and legal effect of a disposition in an instrument is determined by the local law of a particular state selected by the transferor in the instrument. . . .”].) In construing a trust we consider the law in existence at the time it was created. A settlor and his attorney are presumed to know the law at the time the trust is drafted. (Sefton v. Sefton (2012) 206 Cal.App.4th 875, 888 [ 142 Cal.Rptr.3d 174 ]; Ehrenclou v. MacDonald (2004) 117 Cal.App.4th 364, 370 [ 12 Cal.Rptr.3d 411 ].) With these principles in mind, we are not persuaded by the legal authority on which the court relied.

(1) Sections 16340 and 16341

The court first pointed to sections 16340 and 16341, part of the Uniform Principal and Income Act. (§ 16320 et seq.; Act.) Section 16340, subdivision (d) provides that after making specific gifts described in subdivision (b), the trustee must distribute net income as set out in section 16341.

Section 16341 instructs that the beneficiaries described in section 16340, subdivision (d) are entitled to payment of “net income equal to the beneficiary’s fractional interest in undistributed principal assets, using values as of the distribution dates and without reducing the values by any unpaid principal obligations.” (§ 16341, subd. (a).)

The court found there was no “contrary provision” in the Trust and that therefore, under this subdivision, “the fractional interests of the residuary [B]eneficiaries in income follow their fractional income in principal.” This, the court stated, “is the changing fraction method as applied to income.” We cannot agree.

As detailed above, there are contrary provisions in the Trust, and, as Cathe points out, section 16341, subdivision (a) applies only when the assets to be distributed to more than one beneficiary are held in a single pool, not the case here. Rather, under the Trust, once the Subtrusts are created and the interests in the residuary assets are vested, the assets are divided into and held in three separate funds with one beneficiary each.

Moreover, the terms of the Trust and applicable law prohibit application of sections 16340 and 16341. Section 16335, subdivision (a) provides: “In allocating receipts and disbursements to or between principal and income, and with respect to any other matter within the scope of this chapter, a fiduciary: [¶] (1) Shall administer a trust or decedent’s estate in accordance with the trust or the will, even if there is a different provision in this chapter.”

The Trust provided that unless otherwise stated therein, “the determination of all matters with respect to what is principal or income of the trust estates and the apportionment and allocation of receipts and expenses between these accounts shall be governed by the provisions of the Revised Uniform Principal and Income Act of California from time to time existing.” (Art. V, § E, par. 18.) Nothing beyond that is left to the Act.

Katy’s and Lucky’s arguments that the Act is not preempted by the Trust are not persuasive. The Trust has numerous provisions discussing distribution of the assets. As shown above the Trust provides for creation, vesting, and funding of the three Subtrusts upon Donald’s death. The Trust described how each Subtrust was to distribute net income and principal, in general, income in quarterly or more frequent payments (Art. IV, §§ C, par. 6a, D, par. 2c, E, par. 2c) and principal payments for health, education, support, or maintenance. (Art. IV, §§ C, par. 6b, D, par. 2d, E, par. d.)

The Trust also allowed preliminary payments before the end of the closing period so long as they were in equitable amounts. (Art. V, § F, par. 13.) And if non-pro-rata allocations were made to the Subtrusts, the assets must have had a proportionate fair market value. (Art. V, § E, par. 17 a.)

Because the Trust required the division of the residuary into the Subtrusts there is no need to resort to the Act to control non-pro-rata distributions of principal, contrary to the statement of decision.

In addition to finding the changing fraction method applied to interest, the court, while acknowledging section 16341 did not apply “explicitly” to principal, decided the section “plainly recognize[d] both the possibility that fractional interests in principal can change, and the need to allocate receipts to the beneficiaries on the basis of their changed fractional interests in undistributed principal assets, rather than any original or prior interests therein.” From this, the court concluded that although “the drafters of . . . [section] 16341 may not have recognized a need to address the question of principal receipts explicitly, absent clear guidance otherwise, the Court sees no reason in law or in equity to give [principal receipts] special treatment.” This conclusion is flawed.

As stated above, section 16341 applies only to an undifferentiated pool of assets, unlike those here. Second, in construing a statute, we look to the plain language of the statute and where it is not ambiguous “ ‘no court need, or should, go beyond that pure expression of legislative intent.’ [Citation.]” (Ruiz v. Sylva (2002) 102 Cal.App.4th 199, 209 [ 125 Cal.Rptr.2d 351 ].) “Also, where the language of the statute is clear in itself, the court should refrain from artificially adding to or altering it. (See generally, Code Civ. Proc., § 1858 [‘In the construction of a statute or instrument, the office of the Judge is simply to ascertain and declare what is in terms or in substance contained therein, not to insert what has been omitted, or to omit what has been inserted . . . .’].)” (California School Employees Assn. v. Azusa Unified School Dist. (1984) 152 Cal.App.3d 580, 594 [ 199 Cal.Rptr. 635 ].)

Section 16341 is not ambiguous. Therefore, we must look to its plain language, which does not address principal receipts. We see no basis to add to the statute a meaning not contained therein. It is not at all clear from its language. And we are not convinced the drafters did not consider the issue of principal receipts or that they intended the statute to apply to principal.

(2) Out-of-state Cases, Treatises, and IRS Authority

The court cited, without discussion, four out-of-state cases, two law review articles, and one treatise. Contrary to the court’s description, we do not consider this “substantial authority.” This is the only authority any of the parties has been able to find on the matter, and none of that authority is binding.

Not one of these authorities discusses or applies California law. “Where out-of-state authority is at odds with California law, it lacks even persuasive value. [Citation.]” (Lucent Technologies, Inc. v. Board of Equalization (2015) 241 Cal.App.4th 19, 35 [ 193 Cal.Rptr.3d 323 ].) Further, the cases were all based on an ambiguous will or trust, or in one case intestacy, where the court had to resort to the changing fraction method in the absence of any other provisions or evidence of the trustor’s intent.

As to the articles and treatises, they certainly are not binding (Brady v. Calsol, Inc. (2015) 241 Cal.App.4th 1212, 1225, fn. 3 [ 194 Cal.Rptr.3d 243 ]), and Katy and Lucky have not shown they applied the changing fraction method in a fact situation similar to the one here.

Finally, the court’s reliance on an Internal Revenue Code section and a related Treasury regulation is not well taken. Again, this authority, even if it does apply to the interpretation of a trust, deals with situations where no separate trusts are created. ( 26 U.S.C. § 63 ; 26 C.F.R § 1.663(c)-5 (2015).)

d. Equity Is Not a Basis for Using the Changing Fraction Method.

The court found equity in general and as applied to the specific facts of this case supported use of the changing fraction method. The court observed it was only fair a person with a bigger stake in the pool of assets be entitled to same fractional gain or loss on the assets. But these findings too were based on the mistaken presumption of one undifferentiated pool of assets as opposed to three separate Subtrusts.

The court pointed to the Trustees’ and Barcal’s correspondence in which they “appli[ed] a rudimentary form” of the changing fraction method, without calling it such, based on their experience as CPA’s. This was the basis for the May 2012 letter, signed by all the Beneficiaries. This showed the fractional interests of 46.795 percent for Katy, 27.042 percent for Cathe, and 26.163 percent for Lucky. But the court specifically found neither the letter nor the signatures had any binding effect and were merely “meaningful evidence” of the equity of using the changing fraction method.

Lucky points to a letter written by Barcal in which he claims Barcal stated distributions in the 50-25-25 percent ratio were logical and supported by the Trust. What the letter stated, however, was that the “allocation of estate taxes” in these proportions was “logical and is supported by the [TJrust.” In reference to the percentages set out in the tax return, Barcal testified, “Those are the [estate tax] values .... It is not the final probate trust accounting.” He went on, “There are three different types of accountings, and this one is fine as long as you recognize it is only [an estate tax] accounting. That is all it is intended for.”

Barcal’s letter also stated a division of the assets, one-third each, was “logical and is supported by the [T]rust.” Moreover it is clear from that letter Barcal was attempting to resolve the dispute among the parties. This does not in any way support the use of equity to violate Donald’s intent.

e. Expert Testimony

The court made brief reference to the testimony of the experts, noting Katy’s favored the changing fraction method while Cathe’s did not. The court pointed to testimony of Cathe’s expert that the changing fraction method should not be used here, but he would use it in the case of appreciating stocks. That testimony was based on a hypothetical where there were no separate subtrusts. That is not the case here.

Furthermore, we are charged with interpreting the Trust, which is a legal question, not appropriate for expert opinion. (Devin v. United Services Auto. Assn. (1992) 6 Cal.App.4th 1149, 1157, fn. 5 [ 8 Cal.Rptr.2d 263 ].) Thus, the expert opinion evidence does not justify application of the changing fraction method.

f. Conclusion

In sum, the Trust does not support use of the changing fraction method. The Trust makes clear the residuary assets were to be divided on Donald’s death one-third each. The overall intent of the Trust demonstrates that when the Subtrusts are created the assets are to be divided into each Subtrust and administered as such. Even if this need not be done for an initial six-month period, it is consistent with the Trust’s intent that while the assets are marshaled, they be treated as if divided on Donald’s death.

There is no language in the Trust stating that when taxes were charged to Cathe and Lucky or non-pro-rata distributions were made, the fractional shares of the Beneficiaries were to change.

If all of the Trust provisions cannot be completely harmonized, and even if some provisions, read in isolation, might seem to support the changing fraction method, what is clear is that the Trust as a whole does not support its application. “ ‘ “In construing a trust instrument, the intent of the trustor prevails and it must be ascertained from the whole of the trust instrument, not just separate parts of it.” ’ ” (Estate of Cairns, supra, 188 Cal.App.4th at p. 944 .) “ ‘Once the testamentary scheme or general intention [of a trust] is discovered, the meaning of particular words and phrases is to be subordinated to this scheme, plan or dominant purpose . . . .’ ” (Estate of Goyette (2004) 123 Cal.App.4th 67, 73 [ 19 Cal.Rptr.3d 760 ].)

Nor are any of the other bases on which the court relied sufficient to justify use of the changing fraction method. The judgment must be reversed.

4. Taxes on the Goldenrod Property Should Not Have Been Charged to Cathe.

The Trust awarded the Goldenrod Property to Cathe and provided it “shall be free from any federal estate taxes, inheritance or succession taxes, generation skipping transfer taxes, or any other taxes payable upon or resulting from or by reason of the death of the settlor.” (Art. IV, § B, par. 2b, h.) Nevertheless, the court ruled Cathe should be responsible for one-half of the estate taxes on the Goldenrod Property. This ruling must also be reversed.

The court construed that provision to mean Cathe would not be charged with the “full portion” of estate taxes. Rather, she should pay no more or no less than what would have been owed had the Goldenrod Property been given to a third party. The court stated there was no evidence Donald intended to “give Cathe a greater share in the residue of the Trust than Lucky. Accordingly, no adjustment shall be made as between Cathe’s and Lucky’s interests in the residuary portion of the Trust in connection with estate taxes on Goldenrod.” This was error.

The Trust itself unqualifiedly states Cathe should not bear any of the estate taxes on the Goldenrod Property. Once more, we must construe the Trust according to its plain meaning and Donald’s intent. (Brown v. Labow, supra, 157 Cal.App.4th at p. 812 .)

Granted, the Trust mandated Cathe and Lucky were to bear the burden of the estate taxes. (Art. IV, § C., par. 2; Art. VIII, § A.) But the paragraph instructing Cathe should bear no estate taxes on the Goldenrod Property is plain, specific, and unqualified. It stated her gift “shall be free” from “any” taxes. Moreover, the Trust provided the “direction to prorate and apportion [taxes] shall not apply to any distribution which is specifically referred to as being free from taxes.” (Art. VIII, § A.) “[U]nder well established principles of contract interpretation, when a general and a particular provision are inconsistent, the particular and specific provision is [paramount] to the general provision. [Citations.]” (Prouty v. Gores Technology Group (2004) 121 Cal.App.4th 1225, 1235 [ 18 Cal.Rptr.3d 178 ].) Thus the general provision requiring Cathe and Lucky to pay the estate taxes must defer to the specific provision that Cathe pays no taxes on the Goldenrod Property.

Lucky argues it is “startling and counterintuitive” that Cathe’s share of the residuary would bear less of a tax burden than his “not just despite” but “because” of her additional specific gift of the property. And as a result, he will receive a smaller portion of the residue. But, interpreting the Trust provisions as written, such is the result. And there is nothing in the Trust to support any other result. We do not find it “bizarre,” and even if it were that does not trump the Trust provisions, which control. The Trust does not show any intent Cathe should bear one-half or any portion of the estate taxes on the Goldenrod Property.

5. Katy’s Cross-appeal

Katy raises three issues on her cross-appeal, all having to do with the court’s application of the changing fraction method. First, she challenges the court’s revaluation of the assets after values were established for the federal estate tax return. Second, she disputes how the court applied an income tax overpayment. Finally, she claims the court erred in excluding more than $2.5 million attributable to her life estate in several pieces of residential real property from the value of the residue.

Because we are reversing the court’s decision that the changing fraction method should be applied and remanding the case back for the trial court to give new instructions about allocation and distribution of the residuary, these claims are moot. They are all encompassed within the new order the court will issue.

DISPOSITION

We reverse the judgment and remand the matter back to the trial court. The trial court should instruct the Trustees to distribute the residue one-third each to the Beneficiaries after making proper adjustments for distributions, payments, and tax prorations as required by the Trust and the Probate Code.

Further, the trial court should issue new instructions to the Trustees to refrain from charging to Cathe any federal estate taxes, inheritance or succession taxes, generation-skipping transfer taxes, or any other taxes payable upon or resulting from or by reason of the death of the settlor on the Goldenrod Property.

Cathe is entitled to costs on appeal.

This was a complete restatement of the Trust originally executed in 1977. (Art. I.) All references to articles are to the Trust unless otherwise indicated. Lucky is represented in this action by a trustee ad litem who is actually the party. For convenience and clarity, however, we refer to him as Lucky. And, because all of the Beneficiaries have the same last name, we refer to them by their first names, also for clarity, not out of disrespect. Although application of the changing fraction method will result in a reduction of Lucky’s distribution from the Trust, Katy has promised she will leave all of her estate to him upon her death. For the years 2009 through 2012 the Trust was being paid on average $1.2 million per month. Lucky paid slightly more of the estate tax because the Goldenrod Property was given to Cathe tax-free. (Art. IV, § B, par. 2b.) He was not qualified as an expert at trial and participated in the trial as counsel for Lucky. This was the court’s description of the accounting and distribution method based on a constant one-third division to each Subtrust. It also listed Estate of Libeu (1988) 205 Cal.App.3d 1436 [ 253 Cal.Rptr. 456 ], which does not mention or discuss the changing fraction method. The language of the statement of decision is unclear as to when the court determined the Subtrusts were to be funded. Although it talks about funding the Subtrusts, it also discusses the equity of allocating receipts, gains, and losses depending “on the stake that each party has in the underlying asset pool that produced such receipts and gains (or loss).” The court also considered non-pro-rata distributions to the Beneficiaries during the Trust administration, finding they also required adjustments to the proportional interests. But just as with the charges for estate taxes, this would not have required any adjustments. In addition, the Trust provides “[a]ny distribution or division may be made without regard to differences in tax bases of any such property.” (Art. V, § E, par. 17a.) Goren, Cathe’s expert, referred to this as “truing up.” As of December 2012, Katy had been paid almost $40 million, and, in addition to the $19.4 million each paid to Cathe’s and Lucky’s Subtrusts, they had been paid approximately $21 million and just over $18 million, respectively. Although the May 2012 letter stated Barcal had reviewed it, Barcal testified he did not see it until a few months after it was sent.