Gary G. TANGHE, Appellant,
v.
Jackie D. TANGHE, Appellee.
No. S-11222.
Supreme Court of Alaska.
June 24, 2005.*568 Bruce A. Bookman, Bookman & Helm, Anchorage, for Appellant.
Karla F. Huntington, Anchorage, for Appellee.
Before: BRYNER, Chief Justice, MATTHEWS, EASTAUGH, FABE, and CARPENETI, Justices.
OPINION
MATTHEWS, Justice.
The main issue in this case is whether the marital and separate property components of a 401(k) plan should be determined by using a coverture fraction or by tracing the earnings of the separate property component. We conclude that the latter method should be used because it is accurate and the former is not. A second issue is whether survivor benefits payable under a qualified domestic relations order (QDRO) must be capitalized based on the longer life expectancy of the non-employee spouse. We conclude that *569 they need not be capitalized because doing so places an unfair risk on the non-employee spouse and is inconsistent with the "wait and see" approach underlying the use of QDROs.
Gary and Jackie Tanghe were married on April 4, 1992, and they separated on September 13, 2002. During the period of coverture they were both employed by CSX. After they separated Gary took early retirement. At the time of trial Jackie was forty-seven years old and was earning a salary of approximately $90,000. Gary was ten years older and despite taking early retirement was still employed and earning a salary of some $100,000.
Gary challenges a number of aspects of the superior court's property division.[1] We determine most of them summarily. But the two issues noted above require discussion. We turn first to the question of capitalization of survivor benefits.
Capitalization of Survivor Benefits
Gary and Jackie both have defined benefit pension accounts through CSX. Because the parties were still married when Gary retired from CSX, Gary was required either to obtain Jackie's consent to a waiver of benefits or to designate Jackie as his surviving spouse to receive at least fifty percent of what Gary would receive at his death. Gary elected the fifty percent survivor option for Jackie and Jackie agreed to elect a fifty percent survivor option for Gary when she retired. Both pensions were divided by the trial court into marital and separate portions and, under court-ordered QDROs, the marital portions will be equally divided as they are paid. This arrangement is not contested.
Gary argues that according to actuarial tables Jackie will live 12.4 years longer than he. If reality matches the tables Jackie will receive half of Gary's pension, some $1,409 per month, after he dies. She would also receive during that period half of the marital value of her own pension representing the amounts that were previously being paid to Gary. According to an expert witness, when capitalized, Jackie's survivorship interest in Gary's pension is worth about $52,000 and her reversionary interest in her own pension after the projected date of Gary's death is worth about $13,000. Gary argues that the court should have assigned values for these interests to Jackie's side of the ledger.
The trial court rejected Gary's argument, finding "the eventuality of Ms. Tanghe receiving the benefit too speculative to include in the marital estate."
Gary relies on the cases of Zito v. Zito[2] and Broadribb v. Broadribb[3] for the proposition that spouses are presumptively entitled to survivor benefits because they "are an intrinsic part of `the retirement benefits earned during the marriage.'"[4] That proposition is not in doubt. But the question presented is whether it is error to decline to value contingent survivorship benefits that are included in a QDRO. As to this question Broadribb offers no guidance. The survivor benefit there was not contingent. It was "the equivalent of a vested life insurance policy"[5] and it relieved the husband from the need to buy other insurance on his life. Zito is more relevant, but it does not help Gary. *570 There we held that a QDRO dividing the marital share of the husband's retirement benefits should also have included survivorship benefits in case the husband died before the wife.[6] This is consistent with what was done in the present case. But we did not suggest that the value of the benefits in Zito should have been capitalized and credited to the wife's account in addition to being included in the QDRO.
The income streams in the present case will have value for Jackie only if Gary dies before she does. This type of survivor benefit resembles a nonvested pension because one party bears all the risk that the benefit may never be realized. In Laing v. Laing, we rejected the capitalization method for nonvested pensions.[7] "Since the non-employee spouse receives his or her share in a lump sum at the time of divorce, the method unfairly places all risk of possible forfeiture on the employee spouse."[8] Similarly, in the present case, the capitalization method would place the risk of not outliving Gary, or not outliving him by 12.4 years, on Jackie.
Gary's argument that the survivorship benefits be capitalized is inconsistent with the QDRO method of distribution adopted by the court. The QDRO method does not require the valuation of funds being distributed. Unlike the capitalization method advocated by Gary, which requires the use of discount rates and mortality tables, the QDRO method simply links distributions to events as they occur.
Gary is correct in suggesting that what may at first glance appear to be an equal division of pension benefits under a QDRO might actually be unequal because of the differences in life expectancy of the parties. That seemed to be the case in Nicholson v. Wolfe,[9] where the older husband argued that his share of the wife's pension should be capitalized and paid to him in a lump sum rather than in a QDRO because he would probably not live long enough to receive much in the way of benefits from the QDRO. We upheld the QDRO method of distribution in that case.[10] Among our reasons was the fact that the risk of not benefitting from his wife's pension was the same risk that the husband had faced during the marriage, given the parties' age differences.[11] This risk was not increased by the QDRO.
Here, as in Nicholson, the use of QDROs did not alter the parties' pre-divorce chances of benefitting from their spouse's pensions. That may not be a complete answer to Gary's argument that he has gotten the short end of the division of the parties' pensions. But it is good enough, in our view, when combined with the convenience of the QDRO method and the potentially unfair risk that capitalization would impose on Jackie, to lead us to conclude that the court did not abuse its discretion in declining to capitalize the survivor benefits that Jackie may receive under the QDRO.
Determination of the Marital Portion of 401(k) Plans
Both Gary and Jackie contributed to 401(k) plans through CSX before and during the marriage. At the time of trial the balance in Gary's account was $648,473 and the balance in Jackie's account was $302,548. Gary began contributing to his plan in February of 1982 and Jackie started her plan a month later.
Early statements for Jackie's 401(k) plan were missing. The earliest documents available were issued one and one-half years into their marriage. Gary had written down information from the missing statements in order to calculate the growth of each investment over time. He testified that he returned the originals to Jackie's son, but Jackie claimed she could not find them. While the originals were missing at the time of trial, Gary's written record of them was available. The trial court found that Gary had a total of 256 months in his plan, Jackie had 254 months in hers, and that both parties *571 had 125.5 months of marital participation. The court prorated the funds according to the resulting coverture fractions. Thus, 49% of Gary's plan and 49.4% of Jackie's were found to be marital property. Applying these percentages to the current balance of each plan, the court determined $317,751 of Gary's plan to be marital and $149,450 of Jackie's.
Gary takes issue with the court's method. He notes that at the beginning of the marriage he had $212,886 in his 401(k) account. This was separate property, and the earnings on this amount during the marriage were also separate property. He contends that the court should have traced the earnings on the separate property in the account through the years of the marriage. When this is done, the separate funds and their earnings can be subtracted from the total amount in the account in order to yield the portion of the account that is marital property. The court recognized that applying this method would result in Gary's 401(k) having a marital portion worth $174,195, in contrast to the $317,751 allocated by the court when using a straight time proration.
We agree with Gary that in a defined contribution plan, as distinct from a defined benefits plan, proration by funds is the accurate method for distinguishing marital from separate property. Calculating the marital portion of a 401(k) plan through the use of a coverture fraction assumes equal periodic contributions and an equal periodic rate of return. Neither assumption is likely to be accurate. Further, even if these assumptions were accurate, the time proration method would still yield distorted results because it ignores compounding. These are flaws that can result in errors of considerable magnitude. To use this case as an example, the time proration method overstates the marital portion of Gary's account by $143,556.
The leading text on property divisions recognizes that the appropriate method for the valuation of the marital portion of non-defined benefit plans is the proration by funds method:
Proration by funds is always the preferred method for allocating retirement benefits, as proration by time (the basis for the coverture fraction) assumes that the same contributions are made in each and every year of employment an assumption which is untrue in the great majority of cases. Proration by time is used for defined benefit plans only because the complex nature of such plans makes proration by funds impossible to implement.[[12]]
The few cases that have ruled on this subject are generally in accord.[13]
The trial court did not directly disagree with Gary's argument as to how the marital portion of his 401(k) plan should be determined. Instead, the court focused on the fact that the original statements for Jackie's plan were missing for the first one and a half years of their marriage. According to the court, the absence of these statements made accurate use of the proration by funds method as to Jackie's plan impossible. The court concluded that this justified using a time proration method as to both plans.
We do not agree that the absence of a year and a half's information concerning Jackie's plan would justify using a method that yields a manifestly inaccurate result as to Gary's plan. There is no indication that Gary *572 should be charged with responsibility for the lack of information concerning Jackie's plan.
Further, the difficulties in using a proration by funds method for Jackie's plan seem overstated. Gary's written records of the missing statements were not challenged as to accuracy. His records show that the allocations that Jackie made among the seven investment vehicles[14] available under the plan for the years in question, 1992 and 1993, were the same as those she made for 1994, the first year for which original records are available. Further, according to Gary's records Jackie's rate of return for each year was favorable from the standpoint of increasing the separate property component of her plan. In 1992 her rate of return is shown to be about 9%, whereas his was about 8%, and in 1993 her rate of return was about 11%, approximately the same rate that he achieved during that year. Thus there seems to be nothing inherently unreliable about Gary's records.[15]
We conclude that the court erred in dividing the separate from the marital components of Gary's 401(k) plan. We recognize that there may be instances in which so little information is available about the parties' contributions to their 401(k) plans that application of the coverture fraction is warranted,[16] but this is not such a case. On remand the components of the plan should be divided using a funds proration method. With respect to Jackie's plan we believe that the court should also use a funds proration method in order to divide its separate and marital components. In accomplishing this the court may use Gary's records if it finds them credible, or it may use other evidence including reasonable extrapolations from known facts. The superior court should choose the same starting date for each party, even if the court is working with Gary's actual records and extrapolating from Jackie's later records.
Other Issues
Gary raises a number of other issues. We conclude that none of them requires reversal and determine each of them summarily. They are briefly discussed in the margin.[17]
For the above reasons the judgment in this case is affirmed except as to the allocation of the marital and separate property components of the parties' 401(k) plans. With respect to Gary's 401(k) plan the allocation is *573 reversed and with respect to Jackie's plan the allocation is vacated. This case is remanded for further proceedings consistent with this opinion.
AFFIRMED in part, REVERSED and VACATED in part, and REMANDED for further proceedings.
NOTES
[1] The standards under which we review the arguments presented are as follows:
The trial court has broad discretion in fashioning a property division in a divorce action. This court reviews the trial court's determination of what property is available for distribution under an abuse of discretion standard. If in the course of determining what property is available the trial court makes any legal determinations, such determinations are reviewable under the "independent judgment" standard. All questions of law are reviewed de novo with this court adopting the rule of law that is most persuasive in light of precedent, reason and policy. However, the trial court's findings that the parties intended to treat property as marital are disturbed only if clearly erroneous. The valuation of available property is a factual determination that should be reversed only if clearly erroneous. The equitable allocation of property is reviewable under an abuse of discretion standard and will not be reversed "unless it is clearly unjust."
Cox v. Cox, 882 P.2d 909, 913-14 (Alaska 1994) (citations omitted).
[2] 969 P.2d 1144, 1147 (Alaska 1998).
[3] 956 P.2d 1222, 1227 (Alaska 1998).
[4] Zito, 969 P.2d at 1147 (quoting Wahl v. Wahl, 945 P.2d 1229, 1231 (Alaska 1997)).
[5] Broadribb, 956 P.2d at 1227.
[6] Zito, 969 P.2d at 1147-48.
[7] 741 P.2d 649, 657 (Alaska 1987).
[8] Id.
[9] 974 P.2d 417 (Alaska 1999).
[10] Id. at 426.
[11] Id.
[12] See Brett R. Turner, EQUITABLE DISTRIBUTION OF PROPERTY § 6.10, at 523 (2d ed. Supp.2004).
[13] See In re Marriage of Hester, 122 Or.App. 147, 856 P.2d 1048, 1049 (1993) ("When the value of a particular plan is determined by the amount of employee contributions, application of [a coverture fraction] could result in a division of property that is demonstrably inequitable."); Paulone v. Paulone, 437 Pa.Super. 130, 649 A.2d 691, 693-94 (1994) (rejecting the use of the coverture fraction and adopting an accrued benefits test for the distribution of a defined contribution plan); Smith v. Smith, 22 S.W.3d 140, 148-49 (Tex.App.2000) (finding that it was incorrect to apply a coverture fraction to a defined contribution account); Mann v. Mann, 22 Va.App. 459, 470 S.E.2d 605, 607 n. 6 (1996) ("Applying [a coverture] fraction to a defined contribution plan could lead to incongruous results, and such an approach is not generally used."); Bettinger v. Bettinger, 183 W.Va. 528, 396 S.E.2d 709, 718 (1990) (rejecting the use of a discounted present value calculation for division of a defined contribution plan "because no consideration was given to the fact that the fund was earning interest") (quotation marks omitted).
[14] The investment vehicles available consisted of five widely available mutual funds, a guaranteed interest fund, and CSX stock. Their rates of return for the years in question are known.
[15] Further, Gary would have had no reason to suspect that the information he copied down would not be independently verifiable from original statements retained by Jackie or the plan administrator.
[16] See Taylor v. Taylor, 12 S.W.3d 340, 346 (Mo.App.2000) (applying coverture fraction where no records were presented from party's 401(k) plan).
[17] Reimbursement of the Marital Estate for Taxes Paid on Gary's Separate Property. We reject Gary's claim that the trial court erred in requiring him to reimburse the marital estate for some $5,650 in taxes that the marital estate paid on his separate property. Although the court should have made findings on this subject, no remand for findings is necessary because Gary agreed in principle at trial with the proposition that the parties should reimburse the marital estate for marital funds spent on separate assets.
Failing To Credit the Cost of Post-Separation Improvements Made to the Marital Home from Gary's Separate Property. Gary claims that he paid $24,449 in post-separation marital expenses, including some $7,109 used in replacing the deck on the marital home. To pay for these and other expenses he withdrew some $15,000 in marital funds, leaving a balance of $9,449 of separate income that he spent for post-separation expenses. The trial court found that the money he used to rebuild the deck came from the $15,000 withdrawal. This finding is not clearly erroneous. In order to make an effective claim for reimbursement Gary would have to demonstrate a right to reimbursement for all of the claimed post-separation expenditures, rather than just those concerning the deck. He has not attempted this.
Transmutation of the Kenai Cabin. Given the very substantial marital funds and marital efforts that were used to maintain and improve this property, the trial court did not err in concluding that the cabin was properly considered to be marital in character.
Finding that a Fifty/Fifty Division of Property was Appropriate. The record shows that the court considered the factors relevant to the question as to how the marital property should be divided and did not consider any improper factors. The court's conclusion that an equal division of marital property was appropriate was not an abuse of discretion.