x999-1" type="majority">

OPINION OF THE COURT

Wesley, J.

On this appeal we must resolve the issue of how child support should be calculated when parents have “shared custody” of their child. Balancing the policy considerations behind enactment of the Child Support Standards Act (CSSA) against the practical challenges of applying the CSSA in shared custody situations, we hold that child support in a shared custody case should be calculated as it is in any other case. In this case, since the lower courts bypassed the initial three-step statutory formula set forth in the CSSA, we modify and are constrained to remit for a redetermination of child support.

The parties, both practicing attorneys in New York City, were married in September 1986. During their marriage, they had one child, a daughter Morton Elizabeth, born on March 15, 1989. They separated in July 1990 and in February 1992, settled the custody and visitation issues by stipulation. They agreed to a “shared time allocation,” whereby plaintiff (father) would have the child with him from Wednesday evening to Sunday evening one week, and Wednesday evening to Thursday morning the following week.

In April 1993, Supreme Court held a hearing to resolve the issue of child support. Plaintiff then earned $76,876 per year and defendant earned $83,118 per year.

In November 1995, Supreme Court issued a comprehensive opinion in which the court attempted to reconcile the shared custodial arrangement with the requirements of the CSSA (167 Mise 2d 749). The court noted that the statute speaks in terms of a single custodial parent and that “[t]he concept of shared parenting time simply does not appear anywhere in the statute” (id., at 750). Nevertheless, the court concluded that the CSSA “applies” to cases of shared custody (id., at 757).

The court rejected plaintiffs suggestion that it apply a “proportional offset” formula, which would reduce plaintiffs *726 child support obligation based upon the amount of time he spends with his daughter. The court also noted that while other States have adopted various formulas that reduce child support based upon time spent with the child, the CSSA contains no similar formula.

The court then considered how the CSSA should be applied in a shared custody case. After reviewing the CSSA and the economic realities of shared custody, Supreme Court determined that “where there is extensive time sharing the court must look at the totality of the circumstances in both homes rather than rely on the [CSSA] percentages” (id., at 757). The court, therefore, held that “while the CSSA ‘applies’ to shared custody * * * the basic support percentages should not be used in any shared custody case” (id.). The court then applied the factors set out in Domestic Relations Law § 240 (1-b) (f) in the CSSA and fixed plaintiffs basic child support obligation at $750 per month.

The Appellate Division affirmed and noted that “the parties’ unique shared custody arrangement warranted the IAS Court’s resort to the paragraph (f) factors for calculating child support under the [CSSA] * * * in preference to the percentage formula of paragraph (c)” (239 AD2d 106). We granted leave to appeal and now modify.

As a threshold matter, we agree with the lower courts and the parties that the CSSA applies to cases of shared custody (see, Matter of Holmes v Holmes, 184 AD2d 185, 187). The more difficult issue we must resolve is how the CSSA should be applied in cases of shared custody, which in New York encompass a number of situations including joint decision making, joint legal custody or shared physical custody of the child.

The CSSA sets forth “a precisely articulated, three-step method” for determining the basic child support obligation (Matter of Cassano v Cassano, 85 NY2d 649, 652). As we outlined in Cassano:

“[S]tep one of the three-step method is the court’s calculation of the ‘combined parental income’ * * * Second, the court multiplies that figure, up to $80,000, by a specified percentage based upon the number of children in the household — 17% for one child — and then allocates that amount between the *727 parents according to their share of the total income * * *

“Third, where the combined parental income exceeds $80,000 * * * the statute provides that ‘the court shall determine the amount of child support for the amount of the combined parental income in excess of such dollar amount through consideration of the factors set forth in paragraph (f) of this subdivision and/or the child support percentage’ (Id., at 653; see, Domestic Relations Law § 240 [1-b] [c] [1]-[3].) 1

After completing this three-step statutory formula, under the CSSA the trial court must then order the noncustodial parent to pay a pro rata share of the basic child support obligation, unless it finds that amount to be “unjust or inappropriate” based on a consideration of the “paragraph (f)” factors (Domestic Relations Law § 240 [1-b] [f]). Those factors include the financial resources of the parents and the child, the standard of living the child would have had if the marriage had not ended, nonmonetary contributions of the parents toward the child, extraordinary expenses incurred in exercising visitation and any other factors the court determines are relevant (Domestic Relations Law § 240 [1-b] [fD.

Where the court finds the amount derived from the three-step statutory formula to be “unjust or inappropriate,” it must order payment of an amount that is just and appropriate (Domestic Relations Law § 240 [1-b] [g]). If the court rejects the amount derived from the statutory formula, it must set forth in a written order “the amount of each party’s pro rata share of the basic child support obligation” and the reasons the court did not order payment of that amount (Domestic Relations Law § 240 [1-b] [g]).

Plaintiff argues that Supreme Court improperly resorted to the “paragraph (f)” factors, without first determining the basic child support obligation pursuant to the statutory formula. We agree.

The CSSA clearly requires the trial court to first calculate the basic child support obligation, using the three-step statutory formula, before resorting to the “paragraph (f)” factors (Domestic Relations Law § 240 [1-b] [f], [g]). Indeed, even where *728 the trial court rejects the amount derived from the statutory formula, it still must set forth that amount in its written order — “an unbending requirement that cannot be waived by either party or counsel” (Matter of Cassano v Cassano, supra, at 653-654).

The CSSA was the Legislature’s response to the Federal Government’s mandate that States establish mandatory guidelines for determining child support (Matter of Graby v Graby, 87 NY2d 605, 609). This statute replaced a discretionary system and was enacted to “create greater uniformity, predictability and equity in fixing child support awards, while at the same time maintaining a degree of judicial discretion necessary to address unique circumstances” (Matter of Cassano v Cassano, supra, at 652). Here, Supreme Court rejected the Legislature’s chosen method for determining child support, upset the statutory balance struck between predictability and flexibility, and returned to a discretionary approach for a whole class of cases.

Although the CSSA is silent on the issue of shared custody and speaks in terms of a “custodial” and “noncustodial” parent in the application of its methodology, we see no reason to abandon the statute, and its Federally mandated policy considerations, in shared custody cases. While “joint custody” is generally used to describe joint legal custody or joint decision making (see, e.g., Bliss v Ach, 56 NY2d 995, 998; Braiman v Braiman, 44 NY2d 584, 589-590), we are aware that many divorcing parents wish to maximize their parenting opportunities through expanded visitation or shared custody arrangements. However, the reach of the CSSA should not be shortened because of the terminology employed by divorcing parents in resolving their marital disputes and settling custody arrangements. In most instances, the court can determine the custodial parent for purposes of child support by identifying which parent has physical custody of the child for a majority of time (see, Matter of Holmes v Holmes, supra, at 189 [Casey, J., concurring in part and dissenting in part]; see also, Nicholas v Cirelli, 209 AD2d 840; Harmon v Harmon, 173 AD2d 98). As noted by Supreme Court, “[t]he reality of the situation governs” (167 Misc 2d, at 753). Thus, even though each parent has a custodial period in a shared custody arrangement, for purposes of child support, the court can still identify the primary custodial parent.

New York was not a stranger to the concept of shared custody at the time the CSSA was enacted and there were at least two *729 legislative proposals which expressly provided for an adjustment of the child support award in the event parents shared custody of a child (see, 1988 NY Senate Bill 6957-A [court may adjust the pro rata shares of the child support award where parents “share the physical custody of the child”]; 1985 NY Senate Bill 3733-A [court shall adjust child support award where noncustodial parent has physical custody of the child more than 30% of the time]). 2 In light of this legislative history, it is clear that the Legislature considered making an adjustment for shared custody cases, but ultimately rejected the idea and intended the CSSA to apply as adopted. Thus, neither the legislative history nor the statute itself suggests that the Legislature chose to deviate from the initial three-step process in shared custody cases. We can only conclude that the Legislature saw fit not to create an exception to the CSSA for shared custody arrangements, nor will we.

There is no other basis for categorically rejecting application of the statutory formula in all shared custody cases. There will certainly be shared custody cases where the statutory formula yields a result that is just and appropriate, notwithstanding the additional time spent with the child (e.g., the noncustodial parent has the child only 30% of the time and earns substantially more than the custodial parent). 3 Of course, there will also be shared custody cases where the statutory formula yields a result that is unjust or inappropriate. In those cases, however, the trial court can resort to the “paragraph (f)” factors and order payment of an amount that is just and appropriate (see, Domestic Relations Law §240 [1-b] [f], [g]). In addition, application of the formula by the trial court in its initial analysis of the basic child support obligation will facilitate effective appellate review (see, Matter of Cassano v Cassano, supra, at 655). Accordingly, Supreme Court should not have bypassed the three-step statutory formula set forth in the CSSA. The basic framework created by the Legislature can accommodate shared custody cases.

*730 II.

Plaintiff also argues that we should sanction a “proportional offset” formula to bridge the shared custody gap he perceives in the statute. The proportional offset would reduce his child support obligation based upon the amount of time he actually spends with his daughter. Defendant objects to application of the formula, asserting that it is problematic, inappropriate, and has already been tested and abandoned in New York by one intermediate appellate court.

Under the proportional offset formula proposed by plaintiff, each parent’s pro rata share of the basic child support obligation is multiplied by the percentage of time the child spends with the other parent. The two resulting amounts are then offset against each other, and the “net” is paid to the parent with the lower amount (see, e.g., McCauley v McCauley, 172 Misc 2d 611, 615-616). This formula “operates on the theory that each parent owes child support to the other parent based on that parent’s income and the amount of time the child is cared for by the other parent” (Melli and Brown, The Economics of Shared Custody: Developing an Equitable Formula for Dual Residence, 31 Hous L Rev 543, 565).

There are a number of reasons for rejecting the proportional offset formula. First, as we noted earlier, the legislative history and the statute clearly reject this methodology. While the proportional offset formula is currently in use in other States, that is because the Legislatures in those States expressly adopted a formula format (see, e.g., Colo Rev Stat § 14-10-115; Md Fam Code § 12-204; Utah Code Ann § 78-45-2; Va Code Arm § 20-108.2). The difficult policy choices inherent in creating an offset formula for shared custody arrangements are better left to the Legislature.

As Supreme Court noted, the proportional offset formula also fails to account for the generally accepted fact that shared custody is more expensive than sole custody. While it reduces certain costs for the custodial parent, shared custody actually increases the total cost of supporting a child by necessitating duplication of certain household costs in each parent’s home (see, The Economics of Shared Custody, op. cit., at 554; US Dept Health & Human Services Office of Child Support Enforcement, Development of Guidelines for Child Support *731 Orders, at II-59 [1987]; Morgan, Child Support Guidelines: Interpretation and Application § 3.03 [a]). 4

While the over-all cost of supporting a child increases with shared custody, the proportional formula, by offsetting the parents’ child support obligations, can greatly reduce the child support award and deprive the child of needed resources (see, Getman, Changing Formulas for Changing Families: Shared Custody Must Not Shortchange Children, 10 Fam Advoc 47, 49; Development of Guidelines for Child Support Orders, op. cit., at 11-58 [“many practitioners express the opinion that the amounts yielded by guidelines in shared custody cases are inequitable because they are too low”]). Thus, application of the proportional offset formula could potentially undermine one of the primary objectives of the CSSA — to increase child support awards so that children do not “unfairly bear the economic burden of [parental] separation” (see, Governor’s Approval Mem, Bill Jacket, L 1989, ch 567, at 1, 2, reprinted in 1989 NY Legis Ann, at 249-251).

The proportional offset formula generally comes into play only after the noncustodial parent’s time with the child crosses a certain threshold. For example, in Vermont, the noncustodial parent must have physical custody of the child 30% of the time for the formula to apply (Vt Stat Ann, tit 15, § 657). When the threshold is met, there is a “sharp decline” in child support (see, The Economics of Shared Custody, op. cit., at 565; Development of Guidelines for Child Support Orders, op. cit., at II-58). For example, where there is a 30% threshold, a 5% increase in visitation/custodial time may result in a 35% decrease in child support (see, The Economics of Shared Custody, op. cit., at 565). The precipitous drop in child support undoubtedly will encourage a noncustodial parent to seek more custodial time to reduce the child support obligation. In our view, parents should seek shared custody because they desire to spend more time with their children.

The proportional offset formula can also be difficult to apply. The main difficulty is in accurately calculating the percentage of time each parent spends with the child, especially where the parents split time on certain days. In this case, as in most cases, there are days when one parent has the child during the day and the other parent has the child at night. As a result, it *732 is difficult to pinpoint a precise percentage of time that each parent spends with the child. Indeed, the parties hotly dispute the percentage of time plaintiff spends with the child. Plaintiff claims that he spends 42.9% of the time with his daughter, while defendant contends that plaintiff only spends between 32% and 36% of the time with the child.

Finally, the proportional offset formula has the undesirable potential of “encouraging a parent to keep a stop watch on visitation” in order to increase his or her shared custody percentage (Matter of Simmons v Hyland, 235 AD2d 67, 70). Notably, while the Third Department originally endorsed the proportional offset formula (Matter of Holmes v Holmes, 184 AD2d 185, supra), it has reconsidered its prior decision and has found the formula unworkable (Matter of Simmons v Hyland, supra).

For all of the foregoing reasons, we explicitly reject the proportional offset formula. Absent express direction from the Legislature in its precise child support guidelines, we will not reduce the parental resources available to children by applying this problematic formula (see, Matter of Graby v Graby, 87 NY2d, at 613). Shared custody arrangements do not alter the scope and methodology of the CSSA.

The trial court rejected the use of the three-step method for determination of the basic child support obligation as a matter of law. This was error. The matter must therefore be remitted to the trial court for application of the three-step process. If the trial court is satisfied that the amount of basic child support obligation is “unjust or inappropriate” because of the shared custody arrangement of the parents, the court may then utilize “paragraph (f)” to fashion an appropriate award.

We have reviewed plaintiff’s remaining contentions and find them to be without merit.

Accordingly, the order of the Appellate Division should be modified, without costs, by remitting the case to Supreme Court for a redetermination of child support, and, as so modified, affirmed.

Chief Judge Kaye and Judges Titone, Bellacosa, Smith, Levine and Ciparick concur.

Order modified, without costs, and case remitted to Supreme Court, New York County, for further proceedings in accordance with the opinion herein and, as so modified, affirmed.

. Cassano involved Family Court Act § 413, which parallels Domestic Relations Law § 240 (1-b). Both statutes were enacted as part of the CSSA (see, L 1989, ch 567).

. A number of States have adopted a formula for calculating child support when there is expanded visitation/shared custody beyond a threshold (see, e.g., Colo Rev Stat § 14-10-115 [8]; Vt Stat Ann, tit 15, § 657).

. The usual visitation arrangement, if. there is such a thing today, is alternate weekends and one evening a week, plus another two to three weeks for vacations and summer holidays. Over the course of a year, this equates to approximately 20% to 25% of the custodial time with the child (see, Morgan, Child Support Guidelines: Interpretation and Application § 3.03 [a]).

. As Supreme Court noted, some States increase the basic child support percentages by an additional 50% to reflect this increase in total cost caused by duplication (see, e.g., Colo Rev Stat § 14-10-115 [10] [c]).

Citation profile

How this case has been treated — in progress