The University of Iowa College of Law

University of Iowa Legal Studies Research Paper

Number 05-13

April, 2005 Does Parental Autonomy Require Equal Custody at Divorce?

Margaret F. Brinig
College of Law, University of Iowa

An index to the working papers in the University of Iowa Legal Studies Research Paper Series is located at: www.law.uiowa.edu/faculty/workingpapers.php This paper can be downloaded without charge from the Social Science Research Network electronic library at: http://ssrn.com/abstract=708283

Does Parental Autonomy Require Equal Custody at Divorce? Margaret F. Brinig* This paper considers the affect of amendments to state divorce laws that strengthen their joint custody preference. It does so in the context of suits by noncustodial parents challenging substantive custody standards not requiring equal custody at divorce. The complaint is that most custody laws, by using a best interests standard rather than equally dividing custodial time, violate substantive due process. Further, two states, Iowa and Maine, have recently amended their custody legislation to strongly presume joint physical custody. 1 While the Oregon statutes that frame much of the discussion here, like most state laws, do not state an explicit preference for joint custody, 2 shared custody is certainly encouraged by 107.179, 3 which refers cases in which the
*William G. Hammond Professor, Associate Dean for Faculty Development, University of Iowa College of Law.
1

See IOWA CODE 598.41 (2004) (enacted as Acts 2004 (80 G.A.) Ch. 1169 1).

5. a. If joint legal custody is awarded to both parents, the court may award joint physical care upon the request of either parent. If the court denies the request for joint physical care, the determination shall be accompanied by specific findings of fact and conclusions of law that the awarding of joint physical care is not in the best interest of the child. Id. 19A ME. REV. STAT. ANN. 1653 (2001): Laws 2001, c. 329, 2, in subsec. 2, par. D, subpar. (1), inserted: "If either or both parents request an award of shared primary residential care and the court does not award shared primary residential care of the child, the court shall state in its decision the reasons why shared primary residential care is not in the best interest of the child." 2 Joint custody is discussed at some length, and examples provided, in James G. Dwyer, A Taxonomy of Childrens Existing Rights in State Decision Making About Their Relationships, 11 WM. & MARY BILL RTS. J. 845, 911-12 (2003). Dwyer notes that a retreat from joint custody reflects a growing perception that true joint custody, whether physical or legal, though it can be beneficial to children, often is not in a childs best interests, particularly when it is involuntarily imposed on parents and/or when there is a high degree of conflict between the parents. Id. at 911. See generally JUNE CARBONE, FROM PARTNERS TO PARENTS: THE SECOND REVOLUTION IN FAMILY LAW 188 (2000). 3 OR. REV. STAT. 107.179 (2003) provides in part: (1) When either party to a child custody issue, other than one involving temporary custody, whether the issue arises from a case of marital annulment, dissolution or separation, or from a determination of paternity, requests the court to grant joint custody of the minor children of the parties under ORS 107.105, the court, if the other party objects to the request for joint custody, shall proceed under this section. The request under this subsection must be made, in the petition or the response, or otherwise not less than 30 days before the date of trial in the case, except for good cause shown. The court in such circumstances, except as provided in subsection (3) of this section, shall direct the parties to participate in mediation in an effort to resolve their differences concerning custody. The court may order such participation in mediation within a mediation program established by the court or as conducted by any mediator approved by the court. Unless the court or the county provides a mediation service available to the parties, the court may order that the costs of the mediation be paid by one

parties cannot agree on joint custody to mediation, and by 107.105, 4 which

or both of the parties, as the court finds equitable upon consideration of the relative ability of the parties to pay those costs. If, after 90 days, the parties do not arrive at a resolution of their differences, the court shall proceed to determine custody. (2) At its discretion, the court may: (a) Order mediation under this section prior to trial and postpone trial of the case pending the outcome of the mediation, in which case the issue of custody shall be tried only upon failure to resolve the issue of custody by mediation; (b) Order mediation under this section prior to trial and proceed to try the case as to issues other than custody while the parties are at the same time engaged in the mediation, in which case the issue of custody shall be tried separately upon failure to resolve the issue of custody by mediation; or (c) Complete the trial of the case on all issues and order mediation under this section upon the conclusion of the trial, postponing entry of the decree pending outcome of the mediation, in which case the court may enter a temporary decree as to issues other than custody upon completion of the trial or may postpone entry of any decree until the expiration of the mediation period or agreement of the parties as to custody. (3) If either party objects to mediation on the grounds that to participate in mediation would subject the party to severe emotional distress and moves the court to waive mediation, the court shall hold a hearing on the motion. If the court finds it likely that participation in mediation will subject the party to severe emotional distress, the court may waive the requirement of mediation. Id.
4

OR. REV. STAT. 107.105 (2003) provides in relevant part: (1) Whenever the court renders a judgment of marital annulment, dissolution or separation, the court may provide in the judgment: (a) For the future care and custody, by one party or jointly, of all minor children of the parties born, adopted or conceived during the marriage, and for minor children born to the parties prior to the marriage, as the court may deem just and proper under ORS 107.137. The court may hold a hearing to decide the custody issue prior to any other issues. When appropriate, the court shall recognize the value of close contact with both parents and encourage joint parental custody and joint responsibility for the welfare of the children. (b) For parenting time rights of the parent not having custody of such children, and for visitation rights pursuant to a petition filed under ORS 109.119. When a parenting plan has been developed as required by ORS 107.102, the court shall review the parenting plan and, if approved, incorporate the parenting plan into the court's final order. When incorporated into a final order, the parenting plan is determinative of parenting time rights. If the parents have been unable to develop a parenting plan or if either of the parents requests the court to develop a detailed parenting plan, the court shall develop the parenting plan in the best interest of the child, ensuring the noncustodial parent sufficient access to the child to provide for appropriate quality parenting time and assuring the safety of the parties, if implicated. The court may deny parenting time to the noncustodial parent under this subsection only if the court finds that parenting time would endanger the health or safety of the child. The court shall recognize the value of close contact with both parents and encourage, when practicable, joint responsibility for the welfare of such children

requires the court to consider awarding custody jointly. In addition, 1997 Oregon legislation noted in its very first section that it was state policy to [a]ssure minor children frequent and continuing contact with parents who have shown the ability to act in the best interests of the child. The legislation has strengthened the power of noncustodial parents; denial of access to the children would give the right to terminate spousal support, change the parenting plan, or obtain an award for makeup visitation. 5 The legislative history for the bill shows that it was a compromise between mens rights groups and those concerned about domestic violence. 6
and extensive contact between the minor children of the divided marriage and the parties. If the court awards parenting time to a noncustodial parent who has committed abuse, the court shall make adequate provision for the safety of the child and the other parent in accordance with the provisions of ORS 107.718 (4). Id.
5

OR. REV. STAT. 107.434 (2003): provides in part: (2) In addition to any other remedy the court may impose to enforce the provisions of a judgment relating to the parenting plan, the court may: (a) Modify the provisions relating to the parenting plan by: A) Specifying a detailed parenting time schedule; (B) Imposing additional terms and conditions on the existing parenting time schedule; or (C) Ordering additional parenting time, in the best interests of the child, to compensate for wrongful deprivation of parenting time; (b) Order the party who is violating the parenting plan provisions to post bond or security; (c) Order either or both parties to attend counseling or educational sessions that focus on the impact of violation of the parenting plan on children; (d) Award the prevailing party expenses, including, but not limited to, attorney fees, filing fees and court costs, incurred in enforcing the party's parenting plan; (e) Terminate, suspend or modify spousal support; (f) Terminate, suspend or modify child support as provided in ORS 107.431; or (g) Schedule a hearing for modification of custody as provided in ORS 107.135 (11).

One of the fathers rights groups that lists S.B. 243 as a piece of father friendly legislation is (name of group?), http://www.peak.org/~jedwards/fafr97.htm#243 (last visited?) (on file with author).

William J. Howe III, Chair, Oregon Task Force on Family Law, testified on April 4, 1997, before the Senate Business Law and Government Committee. The testimony is reported at http://arcweb.sos.state.or.us/legislative/legislativeminutes/1997%20LEGIS%20WEB/4th%20layer /senate.blg.html/sblg.404.html. He discussed the tension between dads rights groups and domestic violence prevention groups and claimed that the legislation (S. 243) was a compromise.

After setting out the constitutional problem and describing the legislation in some detail, this paper will test the effects of the change in the Oregon statutes. Policy-makers might well want to know how children fare under joint custody as opposed to other possible visitation arrangements. In other words, does the childs best interests, the hallmark of most current statutes, itself require joint custody? Policy-makers might also question whether the stronger legislative preference really increases joint custody awards. Does its requirement that mediation alternatives be suggested (and, in some cases, ordered) in fact increase the number of cases that are settled by mediation? Do judges sometimes prescribe mediation in cases that are inappropriate (such as those in which domestic violence orders have been entered)? Do children receive less child support under the new statutory scheme? Is there evidence that the process makes divorce less painful and less expensive? The broader goal here is to suggest that changes in family law, while often made, are seldom systematically assessed. Society needs such accountability, particularly when children are involved. I will show how it might be done. I. THE CONSTITUTIONAL PROBLEM

The underlying constitutional query is whether the parents rights approach that the Supreme Court has recently found fundamental trumps the best interests test states use in custody litigation. If it does, the current statutes are unconstitutional. The challengers posit that only an equal share of physical custody time will satisfy their constitutional right to direct the upbringing and
During the House Judiciary subcommittee hearings on the same matter, Rep. Michael Fahey introduced HF 3172 on April (date?), proposing that may be changed to shall for joint custody consideration. Rep. Fahey appeared in the show Fathers Issues, Family Issues on October 31, 1996. Faheys testimony on April 3 shows that he believed that unless the judge had hard facts to the contrary, joint custody should be awarded. (Name of document?), at http://arcweb.sos.state.or.us/legislative/legislativeminutes/1997%20LEGIS%20WEB/4th%20layer /house.jud.html/hjudfl.403.html (last visited?) (on file with author). The language changing may to shall was added to SB 243, and now appears as part of 107.105 (1). "It sets clearly a policy that in the absence of abuse or neglect, that it's important for children to have both parents in their lives," said (Jackson County Judge (first name) Orf. "It now becomes the policy of the state." Maya Blackmun, Divorce Laws Aim To Protect Children, THE OREGONIAN, Sept. 28, 1997, at B01. The law also spells out the possible penalties for thwarting visitationsuch as cutting alimony or child support, or ordering makeup visitation time or a hearing to change custodyfor families and judges to better understand. In another article, Jeff Mapes, Lawmakers Put More Parenting In Divorce, THE OREGONIAN, June 25, 1997, at B01, noted: A bill to help both parents have a role in raising their children after a divorce won final approval Tuesday in the Oregon Legislature. Senate Bill 243 would give judges more power to enforce visitation rights for noncustodial parents, even to the point of stopping child and spousal support payments. It also would require divorcing couples to develop parenting plans.

control of their children. 7 So far, the answer given by the courts is that, for a variety of reasons, the parental rights must yield to the childrens. For many years, the Supreme Court has opined that parenting is a fundamental right, that is, a right that cannot be significantly diminished or abrogated without a compelling state interest. 8 Most of the Courts
7

The complaints in each of the fifty suits are approximately the same. For one example, see the complaint in Urso v. Illinois, Urso et al., v. Illinois, #04-CV-6056 (N.D. Ill., Kennelly, J.), dismissed for lack of subject matter jurisdiction (Oct. 7, 2004), pp. 11-12 (cite to which page numbers belong?): 44. The United States Supreme Court has long and consistently held that the care, custody, maintenance, management, companionship, educational choices, and general child-rearing decisions related to one's children are fundamental rights protected by the Federal Constitution. 45. As such, any actions by any person or entity, whether it be by a person acting alone, in conjunction with another, directly or indirectly, by any State entity, or demonstrated by a pattern of deprivations generally attributable to the State itself, that intrude upon these fundamental rights, are patently unconstitutional until, and unless, first validated by a substantially compelling State interest applied with strict scrutiny, and only performed in the least intrusive manner. 46. The State is not permitted to intrude upon these fundamental rights of the natural parent without clear and convincing proof of demonstrable harm to the child(ren) in question. 47. All natural parents existing under the jurisdiction of the State of Illinois are constitutionally entitled to be free of government intrusion in the care, custody, and maintenance of their children, unless there is clear and convincing evidence of proven harm, or of the threat or danger of such harm, to the minor children in question. 48. The United States Supreme Court has consistently reminded that there is a presumption that fit parents act in their children's best interests, and that there is normally no reason for the State to inject itself into the private realm of the family to further question fit parents' ability to make the best decisions regarding their children.

Id. Other examples include Creed et al. v. Wisconsin, 04-00917 (E.D. Wis. Curran, J.), dismissed with prejudice Sept. 24, 2004 (same); Ward et al. v. Louisiana, 04-CV-2697 (E.D. La.) (Fulton, J.); and Martin et al. v. Florida, 04-CV-22385 (S.D. Fla. Jordan, J.). A comprehensive set of links can be found on the website for the Indiana Civil Rights Council, http://www.indianacrc.org/classaction.html (last visited?) (on file with author). The petition for certiorari in Arnold v. Arnold, which the Supreme Court dismissed at 125 S. Ct. 112 (2004), can be found at 2004 WL 1634531. Arnold was an appeal from a denial of a Wisconsin Court of Appeals case reported as Arnold v. Arnold, 679 N.W.2d 296 (Wis. Ct. App. 2004). The Supreme Court of Wisconsin denied review. 691 N.W.2d 354 (Wis. 2005). Arnold raises many of the same issues as do the federal court cases, but in the guise of an appeal from the granting of a 102 day per year partial custody by a state trial court rather than the equal (182.5) days requested.
8

Troxel v. Granville, 530 U.S. 57 (2000) (finding statute violated due process when it allowed visitation petitions to be brought by any one at any time and required parent to

pronouncements have involved state incursions into the province of the parent, 9

demonstrate that visitation was not in the childs best interests); Ankenbrandt v. Richards, 504 U.S. 689 (1992) (stating that though custody and divorce proceedings normally cannot be brought in federal court under diversity jurisdiction, tort suit for abuse by the father could lay in federal court); Michael H. v. Gerald D., 491 U.S. 110 (1989) (holding that even a biological father could not overcome the conclusive presumption of paternity held by the mothers husband over her objection); Mississippi Band of Choctaw Indians v. Holyfield 490 U.S. 30 (1989) (holding that Indian Child Welfare Act required that tribe be allowed to decide dispute between biological Indian and adoptive non-Indian parents); Palmore v. Sidoti, 466 U.S. 429 (1984) (finding that a fit parent cannot be deprived of custody based on racial classifications); Lehr v. Robertson, 463 U.S. 248 (1983) (holding that biological father needed to grasp the opportunity to have a role in his childs life); Santosky v. Kramer, 455 U.S. 745 (1982) (stating that before parental rights could be terminated, state required to show parents unfitness by at least a clear and convincing standard); Lassiter v. Dept. of Social Servs, 452 U.S. 18 (1981) (holding that because of the importance of the relationship, the state could not terminate the rights of mother without providing her counsel in complicated cases); Stanley v. Illinois, 405 U.S. 645 (1972) (holding that a state could not conclusively presume that an unwed but involved biological father was unfit to have custody of his children after their mothers death); Wisconsin v. Yoder, 406 U.S. 205 (1972) (holding that parents right to direct and control the education of their children meant that Amish parents who offered a long tradition of education in domestic arts and farming were exempt from compulsory education statutes for the two years following elementary school); Levy v. Louisiana, 391 U.S. 68 (1968) (holding that a state could not make status to sue for wrongful death of a parent depend upon whether biological parents had married); May v. Anderson, 345 U.S. 528 (1953) (finding that because of the importance of custody, state could not award custody to parent without personal jurisdiction); Prince v. Massachusetts, 321 U.S. 158 (1944) (stating that though parent is free to direct the upbringing of her child, she cannot claim the first amendment allows the child to become a martyr by violating child labor and hour laws); Yarborough v. Yarborough, 290 U.S. 202, 220-21 (1933) (Stone, J., dissenting) (stating that a child living with grandmother could not bring suit against father in federal court based on divorce agreement with mother that he had honored); Pierce v. Society of Sisters, 268 U.S. 510 (1925) (holding that compulsory education statutes unconstitutionally limited permitted schooling to public schools); Meyer v. Nebraska, 262 U.S. 390 (1923) (holding that a state forbidding teaching of modern foreign languages impermissibly interfered with teachers occupation and parents right to direct childrens education). See also Gonzales ex rel. Gonzales v. Reno, 212 F.3d 1338 (11th Cir), cert. denied, 120 S .Ct. 2737 (2000) ([T]he INS policyby presuming that the parent is the sole, appropriate representative for a childgives paramount consideration to the primary role of parents in the upbringing of their children. But we cannot conclude that the policy's stress on the parent-child relationship is unreasonable. See Ginsberg v. New York, 390 U.S. 629, 88 S. Ct. 1274, 1280, 20 L. Ed.2d 195 (1968).). For an analysis of the early cases, see Barbara Bennett Woodhouse, Who Owns the Child? Meyer and Pierce And The Child As Property, 33 WM. & MARY L. REV. 995 (1992) (stating that early cases established that parents, as opposed to the state, owned the child). For a discussion of the rights of unwed fathers, see Laura Orem, The Paradox of Unmarried Fathers, 11 WM. & MARY J. WOMEN & LAW 47 (2004); David D. Meyer, Family Ties: Solving the Constitutional Dilemma of the Faultless Father, 41 ARIZ. L. REV. 753 (1999). For a view that the law governing every decision reflects a mix of concerns for the interests of children and adults, see Dwyer, supra note 2, at 845. Holyfield, 490 U.S. at?; Santosky, 455 U.S. at? (parental rights termination); Lehr, 463 U.S. 248 (1983) (terminating unwed fathers parental rights); Stanley, 405 U.S. at? (adhering to presumption that unwed father was unfit to assume custody of children on mothers death); Yoder, 406 U.S. at? (finding state court jurisdiction over adoption by non-Indian of Indian child); Levy, 391 U.S. 68 (1968) (deciding wrongful death statutes treatment of children of unwed parents); Prince, 321 U.S. 158 (enforcing child labor and hour laws).
9

or at least nonparent party interference with the family. 10 Many of the cases have insisted that decisions about custody be made by state courts because an established domestic relations exception to diversity jurisdiction. 11 Furthermore, even federal statutes respect the decisions of other states regarding custody. 12
Troxel, 530 U.S. at 57 (deciding visitation rights of grandparents or any person seeking visitation); Michael H., 491 U.S. at 110 (deciding parental rights of unmarried biological father seeking parental rights of child conceived in adulterous relationship). 11 The Supreme Court denied certiorari in Eichinger v. Eichinger, No. 04-80, 2005 WL 406071 (ruling below, 02C01-9903-DR-259 from the Allen C. Cir. Ct.; affd, Ind. Ct. App. 2004, 31 Fam. L. Rep. 1191 (2/22/05)): Divorced father whose ex-wife was awarded sole custody of their child did not set forth any argument how Ind. Code 31-17-2-8, which authorizes trial court to determine custody and enter custody order in accordance with best interests of child, operates to treat him differently from similarly situated parties, and thus did not show that it violates equal protection; statutes best interest of child standard constitutes compelling state interest that justifies resultant interference with rights of biological parents, and thus does not violate substantive due process. Id. Ankenbrandt, 504 U.S. 689, 703, 704 (1992) (stating that though a tort suit was permitted: the domestic relations exception, as articulated by this Court since Barber, divests the federal courts of power to issue divorce, alimony, and child custody decrees. Given the long passage of time without any expression of congressional dissatisfaction, we have no trouble today reaffirming the validity of the exception as it pertains to divorce and alimony decrees and child custody orders. Moreover, as a matter of judicial expertise, it makes far more sense to retain the rule that federal courts lack power to issue these types of decrees because of the special proficiency developed by state tribunals over the past century and a half in handling issues that arise in the granting of such decrees. Id. Elk Grove Sch. Dist. v. Newdow, 124 S. Ct. 2301 (2004): One of the principal areas in which this Court has customarily declined to intervene is the realm of domestic relations. Long ago we observed that [t]he whole subject of the domestic relations of husband and wife, parent and child, belongs to the laws of the States and not to the laws of the United States.S. Ct.L. Ed. Id. (quoting In re Burrus, 136 U.S. 586, 593-594, 10 S. Ct. 850, 34 L. Ed. 500 (1890)); see also Mansell v. Mansell, 490 U.S. 581, 587, 109 S. Ct. 2023, 104 L. Ed.2d 675 (1989) ("[D]omestic relations are preeminently matters of state law."); Moore v. Sims, 442 U.S. 415, 435, 99 S. Ct. 2371, 60 L .Ed.2d 994 (1979) ("Family relations are a traditional area of state concern. So strong is our deference to state law in this area that we have recognized a domestic relations exception that divests the federal courts of power to issue divorce, alimony, and child custody decrees.). 12 Although the Parenting Kidnapping and Prevention Act, 28 U.S.C. 1738A (Supp. 2000), requires the states to give Full Faith and Credit to valid foreign custody decrees, the point is to strengthen the effect of a valid first decision of the childs home state. See, e.g., IOWA CODE 598B.102 et seq. (2001) (Uniform Child Custody Jurisdiction Enforcement Act). For one
10

The states, however, have taken radically different approaches to family cases dealing with the same subject matter. While Florida has recently decided that a same-sex couple does not have an equal protection claim that will trump a statute prohibiting them from adopting, 13 courts in states like New Jersey have found that not allowing such couples to adopt would violate state policies against non-discrimination on the basis of sexual orientation. 14 The tradition of deferring to state courts in disputed custody actions began many years ago. For example, in the early nineteen sixties, the Supreme Court decided a case where the parents had been issued conflicting orders by two states, and noted that, left to their own devices, these parents could not make a decision in their childs best interests: Virginia law, like that of probably every State in the Union, requires the court to put the child's interest first. The Supreme Court of Appeals of Virginia has stated this policy with unmistakable clarity: 'In Virginia, we have established the rule that the welfare of the infant is the primary, paramount, and controlling consideration of the court in all controversies between parents over the custody of their minor children. All other matters are subordinate.' Mullen v. Mullen, 188 Va. 259, 269, 49 S.E.2d 349, 354 (1948). Unfortunately, experience has shown that the question of custody, so vital to a child's happiness and well-being, frequently cannot be left to the discretion of parents. This is particularly true where, as here, the estrangement of husband and wife beclouds parental judgment with emotion and prejudice. 15 Consistent with the third party and intrafamily distinction made previously, the Ford Court noted the difference between cases in which outside litigants sought to intrude in matters of family privacy: All of the Virginia cases discussed by the South Carolina court, however, involved purely private controversies which private litigants can settle, and none involved the custody of children where the
states description of the effect of the act, see (title?), http://www.brandeslaw.com/uccjea/an_overview_of_the_uniform_child.htm (New York) (last date visited?) (on file with author); see also People of New York ex rel. Halvey v. Halvey, 330 U.S. 610 (1947) (stating that pre-UCCJEA, it was permissible to modify an out of state order if a change in circumstances occurred); Yarborough v. Yarborough, 290 U.S. 202, 220-21 (1933). 13 Lofton v. Roe, 358 F.3d 804, 807 (11th. Cir. 2004), cert. denied, 125 S. Ct. 869 (2005) (upholding statute prohibiting adoption by same-sex parents despite equal protection claim). Similarly, the Baby M. case was not decided by the Supreme Court despite the tremendous popular interest it generated (title of case? In re Baby M?) 109 N.J. 396, 537 A.2d 1227 (1988). 14 See, e.g., In re J.M.G., 532 A.2d 550, 554 (N.J. Super., Chan. Div., 1993) (citing cases from a number of other states and rejecting an argument of others discrimination quite similar to that made in Palmore, discussed infra). 15 Ford v. Ford, 371 U.S. 187 (1962).

public interest is strong. 16 In those cases involving only custody at divorcein other words, between the two parents themselves, May v. Anderson, 17 Palmore v. Sidoti 18 and Elk Grove Sch. Dist. v. Newdow 19 stand out because the Court necessarily became involved. In each of these, the Court intervened only because there was a constitutional prohibition against what the lower court had done, as in Palmore (where the lower court made a custody determination based on the race of the childs step-parent), or because there was a constitutionally-based jurisdictional defect as in May (where the husband was given custody where the court did not have personal jurisdiction over the wife) 20 and Newdow (where the Court determined that the father, who did not have legal custody over the child in question under a California decision could not challenge the Pledge of Allegiance on her behalf, so the Court ultimately declined to answer the substantive First Amendment question). Otherwise, in all three opinions the Court implied it would not get involved. The Newdow Court, in fact, directly disclaimed the power to second-guess the California courts determination of the underlying custody matter, 21 and deferred to the Eleventh Circuits interpretation of California law. 22 The opinion states:
16 17 18 19 20

Id. at 193. 345 U.S. 528 (1953). 466 U.S. 429 (1984). 124 S. Ct. 2301 (2004). 345 U.S. at 533-34. The Court noted: In Estin v. Estin, supra, and Kreiger v. Kreiger, supra, this Court upheld the validity of a Nevada divorce obtained ex parte by a husband, resident in Nevada, insofar as it dissolved the bonds of matrimony. At the same time, we held Nevada powerless to cut off, in that proceeding, a spouse's right to financial support under the prior decree of another state. In the instant case, we recognize that a mother's right to custody of her children is a personal right entitled to at least as much protection as her right to alimony (footnote omitted).

Id. In Michael H. v. Gerald D., 491 U.S. 110 (1989), the dispute was between one birth parent and the other (and her husband) over custodial rights to the child, Victoria. The Court maintained that the birth father did not have standing in this hopefully untypical situation, in deference to the intact family of the remarried mother. In dissent, Justice Stevens suggested that the childs perspective and rights also needed to be considered. 22 124 S. Ct. at 2311: Newdow's parental status is defined by California's domestic relations law. Our custom on questions of state law ordinarily is to defer to the interpretation of the Court of Appeals for the Circuit in which the State is located. See Bishop v. Wood, 426 U.S. 341, 346-347, 96 S. Ct. 2074, 48 L. Ed.2d 684 (1976). In this case, the Court of Appeals, which possesses greater familiarity with California law, concluded that state law vests in Newdow a cognizable right to influence his daughter's religious upbringing. Newdow II, 313 F.3d, at 504-505. The court based its ruling on two intermediate state appellate cases holding that while the custodial parent undoubtedly has the right to make ultimate decisions concerning the child's religious upbringing, a court
21

One of the principal areas in which this Court has customarily declined to intervene is the realm of domestic relations. Long ago we observed that [t]he whole subject of the domestic relations of husband and wife, parent and child, belongs to the laws of the States and not to the laws of the United States. In re Burrus, 136 U.S. 586, 593-594, 10 S. Ct. 850, 34 L. Ed. 500 (1890). See also Mansell v. Mansell, 490 U.S. 581, 587, 109 S. Ct. 2023, 104 L. Ed.2d 675 (1989) ([D]omestic relations are preeminently matters of state law); Moore v. Sims, 442 U.S. 415, 435, 99 S. Ct. 2371, 60 L. Ed.2d 994 (1979) (Family relations are a traditional area of state concern). So strong is our deference to state law in this area that we have recognized a domestic relations exception that divests the federal courts of power to issue divorce, alimony, and child custody decrees. Ankenbrandt v. Richards, 504 U.S. 689, 703, 112 S. Ct. 2206, 119 L. Ed.2d 468 (1992). We have also acknowledged that it might be appropriate for the federal courts to decline to hear a case involving elements of the domestic relationship, Id., at 705, 112 S. Ct. 2206, even when divorce, alimony, or child custody is not strictly at issue: This would be so when a case presents 'difficult questions of state law bearing on policy problems of substantial public import whose importance transcends the result in the case then at bar.' Such might well be the case if a federal suit was filed prior to effectuation of a divorce, alimony, or child custody decree, and the suit depended on a determination of the status of the parties. Id. at 70506, 112 S. Ct. 2206 (quoting Colorado River, 424 U.S., at 814, 96 S. Ct. 1236). 23 Thus, while rare instances arise in which it is necessary to answer a substantial federal question that transcends or exists apart from the family law issue, see, e.g., Palmore v. Sidoti, 466 U.S. 429, 432-434, 104 S. Ct. 1879, 80 L. Ed.2d 421 (1984), in general it is appropriate for the federal courts to leave delicate issues of
will not enjoin the noncustodial parent from discussing religion with the child or involving the child in his or her religious activities in the absence of a showing that the child will be thereby harmed. Id. (quoting In re Marriage of Murga, 103 Cal. App. 3d 498, 498, 505, 163 Cal. Rptr. 79, 82 (1980); see also In re Marriage of Mentry, 142 Cal.App.3d 260, 268- 270, 190 Cal. Rptr. 843, 849-850 (1983) (relying on Murga to invalidate portion of restraining order barring noncustodial father from engaging children in religious activity or discussion without custodial parent's consent)). 23 124 S. Ct. at 2309.

10

domestic relations to the state courts. The general substantive custody law of each state would seem exactly such a delicate issue, so long as not explicitly based upon race (as in Palmore), or gender (as in the alimony case of Orr v. Orr 24 ). The matters of custody remain firmly subjects of state domestic law. The earlier case of May v. Anderson 25 illustrates both the importance of custody and the individual states interest in children. A husband and wife married in Wisconsin, agreed when they separated in 1946 that the wife should take the children to Ohio to think things over. 26 When she decided not to return, her husband filed for divorce and custody in Wisconsin; the wife was served in Ohio. 27 She never took part in the Wisconsin proceedings and her husband was granted not only the divorce (concededly valid), but also custody; the wife received visitation. 28 After some years under this arrangement, the husband took the children to Ohio to visit the wife where she subsequently refused to return them. The husbands habeas corpus petition was ultimately unsuccessful, with the Supreme Court holding that the ex parte proceeding was not effective to get it [Wisconsin] the personal jurisdiction it must have to deprive [the] mother of her personal right to [the] immediate possession [of her children]. 29 Palmore v. Sidoti 30 is a Supreme Court pronouncement that is perhaps noncustodial parents strongest suit in these legal challenges. There, both divorcing parents were fit custodians, but the mother had been the primary caretaker for the childs entire life. 31 After the divorce, the mother, who was white, married a black man. 32 The Florida lower court worried that continued placement with the Palmores, who lived in a black neighborhood, might disadvantage the child: This Court feels that despite the strides that have been made in bettering relations between the races in this country, it is inevitable that Melanie will, if allowed to remain in her present situation and attains school age and thus more vulnerable to peer pressures, suffer from the social stigmatization that is sure to
440 U.S. 268 (1979) (invalidating an Alabama law allowing alimony awards to wives, but not to husbands). The Louisiana custody case invalidating the use of gender, as opposed to the real life fact is Suire v Jagneaux, 422 So. 2d 572 (La. App. 1982). The Oregon case denying the use of gender alone where all other things were equal is Sweat v. Coughtry, 157 Or. App. 677, 969 P.2d 399, 400 (1998) (holding custody nonetheless went to mother).
25 26 27 28 29 30 31 32 24

345 U.S. 528. Id. Id. 345 U.S. at 530-31. Id. at 534. 466 U.S. at 432 Id. Id.

11

come. 33 The court noted that [t]he judgment of a state court determining or reviewing a child custody decision is not ordinarily a likely candidate for review by this Court. 34 However, the statement about the damaging impact of the neighborhood raises important federal concerns arising from the Constitutions commitment to eradicating discrimination based on race. 35 However, Palmore ultimately will turn out to be unhelpful for plaintiffs in the new class action cases. Despite what plaintiffs would claim about the primacy of parental interests, 36 the Court noted that the court correctly stated that the
Id. Id. at 431 35 Id. at 432. 36 See, e.g., Arnold v. Arnold, 679 N.W.2d 296 (Wis. 2004). Law review articles urging this position, for joint legal custody, include: James W. Bozzomo, Note, Joint Legal Custody: Parents Constitutional Right in a Recognized Family, 31 HOFSTRA L. REV. 547 (2002); Holly L. Robinson, Note, Joint Custody: Constitutional Imperatives, 54 U. CIN. L. REV. 27, 40 (1985) (writing about joint decision-making rather than residential custody). Most states now offer a presumption of joint legal custody if fitness isnt involved. In most situations, an award of joint legal custody will present as direct a threat to the well-being of children as will their joint placement. But see Newdow, 124 S. Ct. at 2310, where Justice Stevens points out the potential harm to plaintiffs child, who seems in a position opposed to that of her father, who challenged under God in the Pledge of Allegiance, and who did not have decision-making authority under the California custody decree:
34 33

This case concerns not merely Newdow's interest in inculcating his child with his views on religion, but also the rights of the child's mother as a parent generally and under the Superior Court orders specifically. And most important, it implicates the interests of a young child who finds herself at the center of a highly public debate over her custody, the propriety of a widespread national ritual, and the meaning of our Constitution. Id. One Note that supports the joint-custody-as-fundamental-right position is Ellen Canacakos, Joint Custody as a Fundamental Right, 23 ARIZ. L. REV. 785 (1981), though she admits that Courts and commentators are unable to agree on exactly what mix of legal and physical custody constitutes the joint custody arrangement, Id. at 787 & n.18, and that the essence of joint custody [is] the sharing, by both parents of responsibility and authority concerning the children. Id. at 788 & n.18. She justifies her position on the grounds that the protection of parental rights derives not from the family unit per se but from the rights of individuals within the family. Id. at 789. Upon divorce, each parent will continue to possess this fundamental right of parental autonomy equally because, even though the family has been dissolved, each parent still retains those family-like bonds recognized by the Court as essential to the right. Id. at 791. This reasoning does not seem to agree precisely with the later position of the Supreme Court in Michael H., 491 U.S. at 123: As we view them, they rest not upon such isolated factors but upon the historic respect--indeed, sanctity would not be too strong a termtraditionally accorded to the relationships that develop within the unitary family. See Stanley, supra, 405 U.S. at 651, 92 S. Ct. at 1212; Quilloin, 434 U.S. at 254-255, 98 S. Ct. at 554-555; Caban, 441 U.S., at 389, 99 S. Ct., at 1766; Lehr, 463 U.S., at 261, 103 S. Ct. at 2993. In Stanley, for example, we forbade the

12

childs welfare was the controlling factor. But that court was entirely candid and made no effort to place its holding on any ground other than race. 37 The court went on to state: The State, of course, has a duty of the highest order to protect the interests of minor children, particularly those of tender years. In common with most states, Florida law mandates that custody determinations be made in the best interests of the children involved. The goal of granting custody based on the best interests of the child is indisputably a substantial governmental interest for purposes of the Equal Protection Clause. 38 The court concluded: The effects of racial prejudice, however real, cannot justify a racial classification removing an infant child from the custody of its natural mother found to be an appropriate person to have such custody. 39 Thus, as a matter of constitutional law, the best interests of the child, protected by the state, should prevail over the constitutional interests of either of the competing parents. Having looked briefly at the constitutional reasons why equal joint placement may not be appropriate, I turn to other, economic arguments. II. AN ECONOMIC APPROACH

The new institutional economics concept that is the most helpful in our discussion of divorcing families is the idea of the externality. 40 Economists, including Ronald Coase, usually describe an externality as an effect on people who were not direct parties to a transaction. 41 Two people are, for example,
destruction of such a family when, upon the death of the mother, the State had sought to remove children from the custody of a father who had lived with and supported them and their mother for 18 years. As Justice Powell stated for the plurality in Moore v. East Cleveland, supra, 431 U.S., at 503, 97 S. Ct., at 1938: Our decisions establish that the Constitution protects the sanctity of the family precisely because the institution of the family is deeply rooted in this Nation's history and tradition. Thus, the legal issue in the present case reduces to whether the relationship between persons in the situation of Michael and Victoria has been treated as a protected family unit under the historic practices of our society, or whether on any other basis it has been accorded special protection. We think it impossible to find that it has. In fact, quite to the contrary, our traditions have protected the marital family (Gerald, Carole, and the child they acknowledge to be theirs) against the sort of claim Michael asserts. 37 466 U.S. at 432. 38 466 U.S. at 433. 39 Id. at 434. 40 . HAL R. VARIAN, INTERMEDIATE MICROECONOMICS: A MODERN APPROACH 545-46 (1993); D. MCCLOSKEY, THE APPLIED THEORY OF PRICE 331 (2d ed, 1985). 41 Ronald H. Coase, in The Problem of Social Cost, 3 J.L. & ECON. 1 (1960), proposed what

13

parties to a contract involving the making of cement. The cement plant is located in a neighborhood, which suffers from the noise, dust, wear, tear and danger presented by the cement trucks. The ill-effects suffered by the neighbors are externalities; much of the law and economics (and economics) literature involves making the parties to such a contract internalize these ill-effects. Sometimes this is done through government regulation, sometimes by giving people in the position of the neighbors the right to sue. Thus the contract may become less than fully enforceable where there are substantial negative third party effects. 42 Although most contracts affect third parties, at least indirectly, sometimes the contracting parties, like those involved in the cement factory transaction, must buy-off the affected outsiders. 43 So long as compensation takes place, 44 the contract remains efficient and enforceable. When the costs to the third party or parties are too high, the contract may be prohibited criminally,45 enjoined, 46 or just not enforced. 47
is popularly called the Coase Theorem. Coases proposition, at least in its incarnation that people should bargain to an efficient outcome regardless of the way the law allocates rights, was first applied to changes in divorce laws in H. Elizabeth Peters, Marriage and Divorce: Informational Constraints and Private Contracting, 76 AM. ECON. REV. 437 (1986) (proposing no increase in rates, but changes in alimony and property distribution). Compare the above with language from Newdow, 124 S. Ct. at 2312: In our view, it is improper for the federal courts to entertain a claim by a plaintiff whose standing to sue is founded on family law rights that are in dispute when prosecution of the lawsuit may have an adverse effect on the person who is the source of the plaintiff's claimed standing. Id. For example, Professor Epstein has written that the only sound justification for inalienability is the practical control of externalities. Richard A. Epstein, Why Restrain Alienation, 85 COLUM. L. REV. 970, 990 (1985) (stating that externalities which are usually present when resources must be shared). (Author?) In Surrogacy: The Case for Full Contractual Enforcement, 81 VA. L. REV. 2305, 2315 (1995), Epstein writes that "the legal response should be to ban or restructure those transactions whose negative third-party consequences outweigh the gains to the transacting parties, . . .[when] gains and losses are measured by a compensation criterion ." See also June Carbone, The Role of Contract Principles in Determining the Validity of Surrogacy Contracts, 28 SANTA CLARA L. REV. 581, 582 (1988) (arguing in favor of enforcement so that genetic fathers can enter the agreements with confidence in the certainty of the outcome). 43 In situations where they do, the contract becomes Pareto-optimal. For discussions of Pareto optimality, see JACK HIRSCHLEIFER, PRICE THEORY AND APPLICATIONS 496-97 (4th ed. 1988); Varian, supra note 35, at 262-63. 44 At least the compensation occurs theoretically, according to Kaldor-Hicks optimality. See, e.g., P.R.G. LAYARD AND A.A. WALTERS, MICROECONOMIC THEORY 32 (1978); Varian, supra note 35, at 218. The original articles are Nicholas Kaldor, Welfare Properties of Economics and Interpersonal Comparisons of Utility, 49 ECON. J. 549-51 (1939); J.R. Hicks, The Valuation of Social Income, 7 ECONOMICA 105-24 (1940). 45 This would include such conduct as drag-racing, which is popularly called chicken. See, e.g., In re Fox, Alleged Delinquent Child, 395 N.E.2d 918 (Ohio Ct Com Pleas 1979). For a game theoretic explanation, see CHARLES GOETZ, LAW AND ECONOMICS 15-17 (1984). 46 This is the case in nuisance situations. See, e.g., Hart v. Wagner, 184 Md. 40, 40 A.2d 47 (1944). 47 I develop this proposition at some length in Margaret F. Brinig, A Maternalistic Approach to Surrogacy, 81 VA. L. REV. 2377 (1995).
42

14

The concept of externalities can be imported into family law. For example, children of divorce suffer substantial negative externalities, 48 so much so, that we may make it more difficult for the parents to divorce. 49 We try to make things easier for these children by requiring child support to the standard of living they would have enjoyed if their parents remained together. 50 In some states, the court may order the divorcing parents to provide the children a college education that children with married parents cannot claim. 51 The difference we have already explored in the constitutional context between the two types of case (where courts may or may not interfere without a showing of parental unfitness) can be explained in economic terms by the presence or absence of third-party externalities. 52 Parents in nearly all divorce cases that involve third-party externalities will be making decisions that affect children. As in some of the post-Troxel grandparent visitation statute cases, divorcing parents may not be acting in their childrens best interests, 53 but rather putting their own needs and interests first. 54
48

As the Newdow Court noted: This case concerns not merely Newdow's interest in inculcating his child with his views on religion, but also the rights of the child's mother as a parent generally and under the Superior Court orders specifically. And most important, it implicates the interests of a young child who finds herself at the center of a highly public debate over her custody, the propriety of a widespread national ritual, and the meaning of our Constitution.

124 S. Ct. at 2310. For similar social science conclusions, see JUDITH S. WALLERSTEIN, SECOND CHANCES: WHAT HAPPENS TO MEN, WOMEN AND CHILDREN A DECADE AFTER DIVORCE (1989); MAVIS HETHERINGTON, MARTHA COX AND ROGER COX, EFFECTS OF DIVORCE ON PARENTS AND CHILDREN, IN NONTRADITIONAL FAMILIES: PARENTING AND CHILD DEVELOPMENT (M.E. Lamb, ed. L. Ehrlbaum, 1982); Mavis Hetherington, Martha Cox and Roger Cox, Long-Term Effects of Divorce and Remarriage on the Adjustment of Children, 24 J. AMER. ACAD. CHILD PSYCH. 518 (1985); Judith S. Wallerstein, The Long-Term Effects of Divorce on Children: A Review, 30 J. AMER. ACAD. CHILD & ADOLESCENT PSYCH. 349 (1991). A less formal account is presented in Barbara Dafoe Whitehead, Dan Quayle was right: Harmful effects of divorce on children, 271 THE ATLANTIC 47 (April 1993). 49 . VA. CODE ANN. 20-91; see also Elizabeth S. Scott, Rational Decisionmaking in Marriage and Divorce, 76 VA. L. REV. 9 (1990); Barbara Defoe Whitehead, A New Familism, FAM. AFF. 1, 5 (1992). 50 Cole v. Cole, 44 Md. App. 435, 409 A.2d 734 (1979); Conway v. Conway, 10 Va. App. 653, 397 S.E.2d 464 (1990). 51 See, e.g., Crocker v. Crocker, 971 P.2d 469 (Ore App 1998); Rohn v. Thuma, 408 N.E.2d 578 (Ind. 1980). Of course, money (and even higher education) does not buy happiness. The requirement that the noncustodial parent pay for college might palliate the pain of divorce somewhat. 52 For a slightly more extended discussion, see MARGARET F. BRINIG, FROM CONTRACT TO COVENANT: BEYOND THE LAW AND ECONOMICS OF THE FAMILY 9 (2000). 53 Dwyer, supra note 2, at 909-25 (suggesting that the standards are sometimes manipulated to simply count along with the rights of parents). 54 See, e.g., Constance v. Traill, 736 So. 2d 971 (La. App. 1999) (finding joint custody inappropriate); Braiman v. Braiman, 378 N.E.2d 1019 (N.Y. 1978) (same); Beck v. Beck, 432

15

Sometimes courts will use this mixed-motives decision-making as grounds for awarding benefits that children of married parents do not enjoy. For example, In re Marriage of Crocker 55 found that although noncustodial fathers paying for their childrens college education were members of a true class, 56 the legislature could rationally decide that divorced parents of college-aged children might be ordered to pay their college expenses: We conclude that the statutory distinction is rational. Even if most divorced or separated parents could cooperate sufficiently to decide whether to support their children attending school, legislators could rationally believe that, because of the nature of divorce and separation, there will be instances in which children will not receive support from their parents to attend school precisely because the parents are divorced or separated, despite the fact that the parents have the resources to provide the support and it is in the children's best interest for them to do so. It might be that, although both parents agree that they should support their child attending school, they disagree on how much each of them should contribute, so that one or both of them contribute nothing. It might be that the nature of the relationship between the parents is so acrimonious that they refuse to agree on anything. It might be that the parent who did not have custody when the child was a minor is unwilling to provide support precisely because he or she did not have custody. It might be that one of the parents who, when married, considered support for his or her child attending school to be a moral obligation, now considers it to be only a legal obligation and, hence, that the parent will provide support only if ordered to do so by a court. 57 Challenges stating that the best interests standards unconstitutionally violate noncustodial parents right to control and direct the upbringing of their child have thus far been unsuccessful. In Arnold v. Arnold, a father challenged a
A.2d 63 (1981) (stating that even though the alternating custody order was affirmed, The necessity for at least minimal parental cooperation in a joint custody arrangement presents a thorny problem of judicial enforcement in a case such as the present one, wherein despite the trial courts determination that joint custody is in the best interests of the child, one parent (here, the mother) nevertheless contends that cooperation is impossible and refuses to abide by the decree). The parties in Beck are still litigating ten years later. See Beck v. Beck, 570 A.2d 1273 (N.J. Super. A.D., 1990); Heinze v. Heinze, 631 N.E.2d 728 (Ill. App. 1994). See generally Vitauts M. Gulbis, Annot. Propriety of Awarding Joint Custody of Children, 17 A.L.R.4th 1013 (1998). Jill Elaine Hasday, in The Canon of Family Law, 57 STAN. L. REV. 825, 849 (2004), argues that despite the rhetoric of best interests, children in custody cases are too often treated like property, though her illustrations are from the termination cases rather than the divorce custody cases. 55 971 P.2d 469 (Or. App. 1998). 56 Id. at 474. 57 Id. at 475.

16

custody award of 102 days a year as opposed to 182.5, claiming that the fundamental liberty interest in equal participation in the raising of his children required equal placement, and thus that the Wisconsin statute merely requiring that participation be regularly recurring and meaningful (but not necessarily equal) was unconstitutional. 58 The court disagreed, finding that he had not shown that the best interests-based statute was unconstitutional beyond a reasonable doubt; the facts of this case are distinguishable from those in Troxel. A dispute between a parent and grandparents represents a far different dynamic than [a] dispute between two natural parents with equal rights after a divorce. 59 III. THE EFFECTS OF JOINT CUSTODY

There is also a practical reason for the best interests as opposed to the higher compelling state interest for divorce custody cases. There are simply too many cases for courts to implement the test,60 while at the same time, the stakes
679 N.W.2d 296 (Wis. 2004), cert. denied, 125 S. Ct. 112 (2004). In the fathers brief supporting his certiorari petition, he claimed (at *15) (citing Troxel): There was no compelling reason to apply "a best interest of the child standard" before a state can intrude into fundamental right to raise his children, of a father in a divorced family, while a similar father's role in an intact family can only be by invaded by a showing of substantial harm. (Need citation of brief) (citing Troxel, (citation, pin cite). Id. at 711 (This page number cant be right). In the Illinois suit, Urso v. Illinois, 04CV6056 (N.D. Ill. 2004), Judge Kennelly dismissed the case, writing:
59 58

Urso's complaint asks this Court to cure a variety of general social problems by revamping Illinois custody law and to rewrite the decisions in literally tens of thousands of Illinois state court custody cases by the stroke of a federal judge's pen, and it therefore does not present such a controversy, as there is no judicially manageable standard by which this Court could resolve the issues his complaint poses. Id. See generally Baker v. Carr, 369 U.S. 186, 217 (1962). See also Eichinger v. Eichinger, Ind. Ct. App. (unpublished), U.S. cert. denied, 04-801 (2005). The circuit court pleadings can be found at http://www.childrensjustice.org/litigation/Litigation/GrassFire_EP/GrassFire_Motions/Indian a/EICHINGER_GrassFire.html (last visited?) (on file with author). The Louisiana class action suit, Ward et al. v. Louisiana, #04-CV-02697, was dismissed by Judge Feldman on March 1, 2005 (E.D. La.) 60 The Iowa judicial branch web site contains the following information: 63.6% of all the (75,615) civil filings in the district court in 2002 were domestic relations cases, including 20,071 dissolutions and modifications, 28055 support actions and 6280 domestic abuse civil filings. In addition, there were 12329 juvenile filings (7022 delinquency, 5234 CINA, 1924 actions to terminate parental rights (49% increase in two years) and 73 other child welfare matters). On the criminal side, there were 6256 indictable domestic abuse cases. On appeal, 47% of the civil cases concerned domestic relations, and child custody issues were involved in over 72% of these.

17

for litigants are extremely high because custody is far more precious than property rights. 61 In fairness to all concerned, one of the empirical questions that should be asked before equal joint custody is presumed is whether it can be shown to be in the childs best interests. 62 If joint custody presents the best situation for children in all (or most) divorcing families, a compelling state interest would support the presumption. Do children in fact fare better, and how would a researcher find out? In the next section of the paper, I tackle these questions from two perspectives. First I will consider the way children of divorce turn out as adolescents depending upon different patterns of overnight stays with their fathers. Second, I will look at the changes in various kinds of costs of divorce once a strong presumption of joint custody is introduced. What would an ideal study of the effect of joint physical custody on children look like? First, a large nationwide random sample of children under various kinds of custody situations, including living with married parents, would be assembled. They would then be followed over time and would be observed and tested regularly along a number of dimensions. 63 This sort of study has not been done before. Instead, a special sample of children of divorce has been asked about their feelings, activities or mental health retrospectively. 64

(Web Site Title) http://www.judicial.state.ia.us/orders/reports/AnnualReport2002amended.doc (last visited?) (on file with author). Of course, if using the test meant that in practice all orders would be 50/50 (since virtually no one could counter the presumption), it would be dramatically easier to implement and would cut down on court time. I argue that this would run counter to the proposition, see Stanley v. Illinois, 405 U.S. 645, 657 (1972), that case-by-case determinations are necessary: Procedure by presumption is always cheaper and easier than individualized determination. But when, as here, the procedure forecloses the determinative issues of competence and care, when it explicitly disdains present realities in deference to past formalities, it needlessly risks running roughshod over the important interests of both parent and child. It therefore cannot stand. In addition, as shown infra, there would likely be tremendous costs engendered by such a change. 61 May, 345 U.S. at 533. 62 This seems to be the tactic that has been used by some courts in deciding against joint custody when the parents were unable to get along. See, e.g., Constance v. Traill, 736 So. 2d 971, 975 (La. App. 1999) (We are not persuaded from the record, viewed in its entirety, that the trial court erred in finding that equal sharing of physical custody between the parents was not in the best interest of these two young girls .); Braiman v. Braiman, 378 N.E.2d 1019, 1021 (N.Y 1978) (It is understandable, therefore, that joint custody is encouraged primarily as a voluntary alternative for relatively stable, amicable parents behaving in mature civilized fashion. As a court-ordered arrangement imposed upon already embattled and embittered parents, accusing one another of serious vices and wrongs, it can only enhance familial chaos.) 63 Even better would be an experimental random assignment to the various custody situations. Obviously this cannot and should not be done in real life. 64 Compare Sanford L. Braver, Ira Mark Ellman, & William V. Fabricius, Relocation of Children After Divorce and Childrens Best Interests: New Evidence and Legal Considerations, 17 J. FAM. PSYCH. 206-19 (2003).

18

I therefore sought a large longitudinal sample of junior high school and high school-aged children. 65 The National Longitudinal Study of Adolescent Health was carried out by the Center for Population Studies of University of North Carolina, 66 and contains many variables of interest including tests and surveys done on children, questions answered by parents and information provided by schools. I restricted the sample to children living with their mothers whose fathers lived elsewhere; a large majority of the parents were divorced. 67 To repeat my goal: a stronger case could be made for plaintiff noncustodial parents position if the children who saw their parents regularly did better . On the other hand, if seeing fathers frequently harmed the children, mothers or the state might claim that joint custody harmed children. The data revealed neither of these possibilities. What I found was that the worst situation for children was when they visited their fathers infrequently. 68 Otherwise, however, there was no increase in custodial time that made a statistically significant difference (See Tables 2-5). The only exception to this rule appears in Table 3, where children
Robert Emery, who is speaking later in the conference, will have more to say about the relationship of joint custody and attachment. See Robert F. Kelly & Shawn L. Ward, Social Science Research and the American Law Institute's Approximation Rule, 40 FAM. CT. REV. 350, 355-59 (2002). 66 Richard Udry & Peter Bearman, The National Survey of Adolescent Health, CAROLINA POPULATION CENTER, UNC (1994-95). The description, found on their website, reads as follows: The National Longitudinal Study of Adolescent Health (Add Health) is a nationally representative study that explores the causes of healthrelated behaviors of adolescents in grades 7 through 12 and their outcomes in young adulthood. Add Health seeks to examine how social contexts (families, friends, peers, schools, neighborhoods, and communities) influence adolescents' health and risk behaviors. Initiated in 1994 under a grant from the National Institute of Child Health and Human Development (NICHD) with co-funding from 17 other federal agencies, Add Health is the largest, most comprehensive survey of adolescents ever undertaken. Data at the individual, family, school, and community levels were collected in two waves between 1994 and 1996. In 2001 and 2002, Add Health respondents, 18 to 26 years old, were re-interviewed in a third wave to investigate the influence that adolescence has on young adulthood. Multiple datasets are available for study, and more than 1000 published reports and journal articles have used the data to analyze aspects of these complex issues. Add Health investigators hope this research will enable policy makers, researchers, health-care providers, and educators to better understand how to protect the health of young people in the US. A description of the research design can be found at http://www.cpc.unc.edu/projects/addhealth/design (last visited?) (on file with author). 67 There were some children living primarily with their fathers, but too few (10) to do credible statistical analysis. Those situations were sufficiently different from the mother-custody ones that using the more common situation seemed preferable. 68 Table 1 shows that 16.8 % of the fathers saw their children several times a year.
65

19

seemed less likely to engage in alcohol, tobacco and marijuana use if they stayed with their father several times a month. A cautionary word about this data (and why it might possibly overstate the seemingly benign effect of joint physical custody) comes from the fact that the custody arrangements in the underlying Add Health study were highly unlikely to have been ordered under presumptive or mandatory equal custody statutes. 69 Since the surveys were taken in the mid-1990s, the most frequent pattern was joint legal custody, with a best interests standard for physical custody. In the sample, only 8% of the children were staying overnight once a week or more with their fathers. 70 A mandatory joint physical custody situation, particularly an equal one, rather than an arrangement worked out by the particular parents in individual cases, is likely to be much less successful. Now we turn to a different look at custody, and present the effects of a change in custody standards to one stressing more equal parenting time, and consider the claims made by the proponents of the Oregon statutes discussed at the beginning of this piece: that requiring the court to consider joint custody, stress parenting plans, provide mediation in contested cases and allow streamlined custody enforcement proceedings will better the divorce process. These claims are testable with other data. Between 1995 and 2002, there were approximately 125,000 divorces in Oregon. 71 As each Oregon divorce was granted, the Circuit Court Clerk sent the
69

We do not even know whether the arrangements reported by the adolescents were those that were court-ordered as opposed to de facto. For similar cautions, see Kelly and Ward, supra note 60, at 360. Kelly and Ward, supra note 60, mention two large studies finding that in the absence of conflict, more frequent contact with noncustodial parents is associated with better psychosocial adjustment of children. Id. at 362 (citing Christy M. Buchanan, Eleanor E. Maccoby & Sanford M. Dornbush, Caught Between Parents: Adolescents Experience in Divorce Homes, 62 CHILD DEV. 1008 (1991); Paul R. Amato & Sandra J. Rezac, Contact with Nonresidential Parents, Interparental Conflict and Childrens Behavior, 15 J. FAM. ISS. 191 (1994)). 70 This is slightly lower than the 12 to 24 percent found only slightly earlier in Joan B. Kelly, The Determination of Child Custody, 4 FUTURE OF CHILDREN 121, 125 (1994). 71 An email from Joyce Grant-Whorley, Oregon Department of Vital Statistics, to the author dated May 18, 2004, shows the following: 1996 1997 1998 1999 2000 2001 2002 14,973 14,880 15,265 15,647 16,583 16,569 16,151

In addition, there were 15,329 in the 1995 records sent to the author. The total is 125,397. I selected the following number of cases each year: 1995 1996 462 490

20

information to the division of Vital Records of the Department of Health and Human Services. This information, more extensive than that collected in most states, 72 included the names, counties of birth and residence of each spouse, their ages, the dates of marriage, separation and divorce, the identity of the plaintiff in the divorce action, the number of the marriage for each spouse, the date and way the previous marriage ended, the education and race of each spouse, the number of minor children in the household, and the custody awarded for each child. I obtained electronic copies of all this information. In addition, I matched each divorce to the Oregon Online Judicial Information Network (OJIN) to obtain specific information about the court proceedings surrounding the divorce. Since 1991, OJIN has collected case information from each countys circuit courts and made it available free of charge at various sites in Oregon, and, for a setup and hourly fee, to online users elsewhere. 73 The OJIN information allowed me to collect data for each case on attorney representation, the number of court incidents (including motions), the amount of fees charged, whether or not a party alleged domestic violence (including whether a protective order was issued), whether or not one alleged failure to pay child support or sought to change visitation or custody, and so forth. First, I randomly selected 500 cases involving children for each of the eight years involved (nearly four years before the statute went into effect in late 1997, and slightly more than four years thereafter). 74 After matching the two electronic databases for the random sample of cases, identifying information was deleted from the files. Each of the approximately 3800 cases was coded for a total of 80 variables, 38 of which came from the divorce certificates.75 After some files were eliminated (because the court (OJIN) records were missing or because neither parent received custody of the children 76 ), data analysis began. Descriptive
1997 1998 1999 2000 2001 2002 438 485 487 457 493 494

The total number of selected cases (with children in the custody of one of the parents) is thus 3806. 72 The only other states to continue to collect as much data, since the National Center for Health Statistics stopped compiling individual divorce data in 1995, are Connecticut, Montana and Virginia. None of these states have online judicial records. 73 Human subject review board permission was given for the matching, based upon names and type of actions that the process required. 74 SPSS, the statistical program, allows a random selection of any given number of cases. There were in excess of 7000 with children each year. 75 All files were cross-checked randomly by a second coder, and those involving some discretion on the part of the law student involved (such as whether or not a domestic violence petition resulted in more than a fleeting protective order, or whether or not mediation actually resolved the case) were reviewed a second time by Brinig. 76 In 166 cases someone other than the parents ended up with custody. This could be because the children were institutionalized, because both parents were incarcerated, or because at

21

variables appear in Table 6. Correlations revealed relationships between several variables of interest, particularly with whether or not the couples separation took place after the statutory revisions went into effect. 77 Note that domestic violence petitions are not related necessarily to when the couple separated, and in fact, may have occurred as early as 1982. The interesting date, therefore, is whether the petitions are filed before or after the statutes effective date in 1997. Similarly, we restricted motions to modify or enforce visitation or change custody to those filed after the divorce. It therefore became important to know when the couple divorced as well as when the motion was filed, and a modified dataset (with a row for each year beginning at the time of divorce and ending when the first custodyrelated motion was filed, up to 2004) was constructed. Table 7 tests whether the strengthening of the joint custody presumption in fact increased joint custody awards. The answer is that it did, by about 30 percent (see last column). Other large coefficients come from longer marriages, marriages with fewer children in the household, and those in which the husband had not married before (perhaps having other children by the prior relationships). Table 8 considers the effect of the change in the custody statutes on the amount of child support awarded. 78 The table shows that separation after the custody statute took effect, holding other things constant, was statistically significantly related to a decrease in the absolute dollars of child support awards, with a difference of about $80 a month. 79 However, even this turns into a larger net loss in buying power for the custodial parent because of inflation during the same time period. The biggest effects (seen in the Standardized Coefficients of Table 8) are (positively) when the wife is represented by an attorney and (negatively) when the population density of the county is high (meaning that higher child support awards were given in rural areas; quite puzzling since the guidelines are supposed to be effective statewide and one would guess that the cost of living would be higher in cities). Although I cannot say for sure whether wives disadvantaged by the new statute were trading money for child custody, it

divorce, neither parent wanted to retain custody of the children and allowed relatives to raise the children. In five cases data about the children was missing from the divorce certificate. 77 According to Oregon law, statutes ordinarily become effective 90 days after the date of passage, which in this instance was July 30, 1997. Although divorce laws and procedures theoretically may be salient before the divorce process, they will likely effect couples behavior during and after the divorce process. The date of separation seemed to be the most approximation we could find in each case. 78 Judgment, unfortunately, cannot be used to address the share of marital property awarded to the wife. More often than not, instead of a property settlement, it reflected a judgment for past due child support. The predicted value of joint custody (from Table 7) was included in the Table 8 regression because joint custody is endogenous, meaning that it will affect the amount of child support but will also be affected by many of the same variables, including the law. There is no better way to combine results from a logit and OLS regression. 79 This holds other things, like joint custody, constant. Considered independently, support decreased from $370.13 before the statute took effect to $341.03 afterward. See Table 10.

22

is perhaps significant that wives who are represented do better. Of course, this may simply be another reflection of the effect of a higher income. Finally, Table 9 considers whether the legislation increased the number of motions to modify or enforce parenting time or child custody. Because these always followed the divorce, the number of years since the divorce was obviously related, and a modified dataset was used that took account for each couple the number of years since the divorce and up to the filing of such a motion (if one was filed, otherwise until 2004) was constructed. The answer is that the number did increase significantly (and almost doubled) following enactment of the statute. Most of these motions were to change custody or visitation, not to enforce parenting time. 80 The motions to modify appeared in nearly 17 percent of the cases, and were, as one would expect, related to the number of children in the household. If the desire of the legislation was to make it easier for unhappy parents to enforce their visitation time, its purpose was clearly not met.81 If it was to aid unrepresented litigants, it failed, since these were importantly and significantly less likely to bring actions. IV. A BROADER LESSON?

As we have seen, child custody rules provide legislators (and noncustodial parents) with tremendous temptation to urge change. Divorce rules generally seem like a zero-sum gameto the extent husbands win, wives lose. But when the rule change affects custody decisions, the possibility for a still greater loss exists for the very children the parents and the state are supposedly protecting. Constitutionalizing child custody, or litigating in terms of individual parents rights, is likely to harm children in many ways. They may end up living with a parent more interested in punishing the former spouse than in doing what the child needs. They may have less money with which to live, as a child support settlement for lower than the guideline amount pays off a parent claiming joint custody, or if a joint custody solution is ordered but not actualized, or if scarce resources are expended on pre or post-divorce litigation. 82 They may live the
Many more disputes probably occurred that were settled by the parties prior to judicial intervention. Sometimes, especially in cases where child support was not involved, there were probably adjustments made to custody arrangements about which the parties never told the court. In total, 11.5% of the petitions were to enforce visitation, 39% were to change visitation (parenting time) and 48.8% were to modify custody. Interestingly, more motions to change custody were filed before the statute than after it became effective (52.9% compared to 45.2%). 81 In divorces before the effective date of the statute, motions to enforce were filed 12.77 percent of the time. After the statute, the corresponding value was 10.4%. 82 The level of congestion formed part of the problems with the system cited by the Oregon Task Force on Family Law, Final Report to Governor John A. Kitzhaber and the Oregon Legislative Assembly, Creating a New Family Conflict Resolution System 4, Dec. 31, 1997 (on file with author) [hereinafter Oregon Task Force Report]: In 1993, more than one-half of circuit court filings statewide were in family/juvenile law, but refer than 20 percent of the courts resources were devoted to this critical area. Similarly, the Annual Statistical Report of the Iowa Judicial Branch July 28, 2003, at http://www.judicial.state.ia.us/orders/reports/ (last visited?) (on file with author), states Civil filings in 2002 totaled 75,61548,126 domestic relations cases and
80

23

life of peripatetic suitcase-dwellers, 83 and even worse, may be shuttled between parents who actively seek to undermine each other. 84 Joint custody may be a fine (and even the optimal) solution if desired by both parents who are willing to work hard towards its success. 85 Mandatory joint custody, or even a movement in that direction, seems to cause a number of other problems that perhaps its proponents did not anticipate. Unfortunately, the biggest winners, at least in Oregon, seem to be not so much the traditionally noncustodial parents, but rather the mediators86 and, slightly less dramatically, the divorce attorneys.

27,489 law and equity matters. 83 One option is to have the parents do the moving. See the testimony of Dr. Shuman in Winn v. Winn, 593 N.W.2d 662, 668-69 (Mich. Ct. App. 1999) (suggesting that ideally the child would live in the home and the parents rote in and out; the court ultimately denied a joint physical custody petition). 84 See, e.g., Murray v. Murray, 2000 WL 827960, *2 (Ct. App. Tenn. 2000): The parties are equally unhappy with the decision of the trial court, and both agree that joint custody is not in the best interest of the children. Interestingly, the trial judge himself stated at the conclusion of the May 12 hearing that there is no way that joint custody is going to continue to work in this case. I don't think it ever really operated or worked, and joint custody is an onerous burdensome method of raising children between divorced people. It rarely really works. It is unclear why the trial judge chose, despite his own grave reservations, to order a joint custody arrangement in this case. Perhaps he ruled as he did because of the difficulty of choosing one parent over another, when both parties appear from the record to be loving, concerned parents, who are obviously eager to do their best for the children. In any case, the parties appear to be in agreement that it would be in the best interest of the children for the court to grant custody to only one parent. Of course they disagree as to which of them is the more suitable parent to exercise that custody. I have argued this before. Margaret F. Brinig, Feminism and Child Custody Under Chapter Two of the American Law Institute's Principles of the Law of Family Dissolution, 8 DUKE J. GENDER L. & POL'Y 301, 314 (2001) (arguing for the replication principle of the ALI Principles of Family Dissolution); see also Elizabeth Scott & Andre Derdeyn, Rethinking Joint Custody, 45 OHIO ST. L.J. 455, 457-58 (1984) (arguing against a presumption of joint physical custody); Margaret F. Brinig & F.H. Buckley, Joint Custody: Bonding and Monitoring Theories, 73 IND. L.J. 393 (1998) (arguing that joint custody may give fathers reasons to work harder on their marriages and more incentives to pay court ordered support). 86 One of the members of the Oregon Task Force whom I corresponded with was Hugh McIsaac. Mr. McIsaac and Charlotte Finn co-authored, Parents Beyond Conflict: A Cognitive Restructuring Model for High-Conflict Families in Divorce, 37 FAM. & CONCILIATION CTS. REV. 74 (1999), in which they describe an education program Mr. McIsaac ran during the time period studied in Multnomah County, Oregon (where Portland is located) for high-conflict families. The abstract notes the authors believed many of the difficulties between parents were caused by the negative perception of the other parent created during the spousal relationship. They also believed the key to successful co-parenting is to frame these perceptions emphasizing cooperation and joint problem-solving. Finally, the authors believed parents must learn to separate conflict in the spousal role from conflict in the parenting role.
85

24

V.

TABLES
Valid Percent 48.6 12.3 16.8 7.3 7 8 100 Cumulative Percent 48.6 60.9 77.7 85 92 100

Table 1. TIMES LAST YEAR KID STAYED WITH BIO DAD Frequency .00 NEVER/NA 1.50 ONCE OR TWICE 4.00 SEVERAL TIMES Valid 10.00 ABOUT ONCE A MONTH 50.00 ABOUT ONCE WEEK 100.00 MORE THAN ONCE A WEEK Total Missing Total 99 1097 277 378 165 158 181 2256 459 2715 Percent 40.4 10.2 13.9 6.1 5.8 6.7 83.1 16.9 100

Table 2. Depression (CESD2 19-item Depression Scale) (R2 (adj.) = .047)

Unstandardized Coefficients B (Constant) KIDAGE AGE AT TIME OF INTERVIEW 1995 KIDSEX HHINC HH INCOME IN 1000S MOMHISP MOM IS HISPANIC MOMBLACK MOM IS BLACK MOMASIAN MOM IS ASIAN MOMNATAM MOM IS NATIVE AMERICAN MOMOTHER MOM IS OTHER RACE MOMEDUC MOTHERS YEARS OF SCHOOLING MOMAGE SELDOM DOES KID STAY W DAD ONCE OR TWICE? 1=Y, 0 = N SEVERAL DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 = N MONTHLY DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 = N WEEKLY DOES KID SEE DAD MORE THAN WEEKLY? 1=Y, 0 = N OFTEN TALKBIOD TIMES LAST YEAR KID TALKED WITH BIO DAD HOWCLOSE HOW CLOSE KID FEELS TO BIO DAD 14.237 0.194 1.318 -0.004 2.148 0.695 0.156 2.979 1.111 -0.311 -0.021 0.671 1.33 -0.024 -0.193 0.278 2.73E-03 -0.595 Std. Error 1.811 0.085 0.278 0.005 2.621 0.311 0.938 1.227 0.605 0.066 0.027 0.444 0.422 0.573 0.607 0.587 0.004 0.124 0.052 0.104 -0.017 0.018 0.05 0.004 0.053 0.04 -0.108 -0.018 0.035 0.078 -0.001 -0.008 0.012 0.018 -0.13 Standardized Coefficients Beta t 7.863 2.272 4.741 0.742 0.82 2.236 0.166 2.428 1.835 4.731 0.784 1.511 3.153 0.043 0.319 0.474 0.671 4.814 Sig. 0 0.023 0 0.458 0.413 0.025 0.868 0.015 0.067 0 0.433 0.131 0.002 0.966 0.75 0.635 0.502 0

Table 3. Drug, Alcohol and Tobacco Use (Times Per Month) (R2 (adj.) =0.80)
Unstandardized Coefficients Standardized Coefficients t Sig.

25

B (Constant) KIDAGE AGE AT TIME OF INTERVIEW 1995 KIDSEX HHINC HH INCOME IN 1000S MOMHISP MOM IS HISPANIC MOMBLACK MOM IS BLACK MOMASIAN MOM IS ASIAN MOMNATAM MOM IS NATIVE AMERICAN MOMOTHER MOM IS OTHER RACE MOMEDUC MOTHERS YEARS OF SCHOOLING MOMAGE SELDOM DOES KID STAY W DAD ONCE OR TWICE? 1=Y, 0 = N SEVERAL DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 = N MONTHLY DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 = N WEEKLY DOES KID SEE DAD MORE THAN WEEKLY? 1=Y, 0 = N OFTEN TALKBIOD TIMES LAST YEAR KID TALKED WITH BIO DAD HOWCLOSE HOW CLOSE KID FEELS TO BIO DAD -0.685 0.317 -0.148 -0.004 -1.783 -1.946 -1.302 -1.48 -1.172 -9.07E-02 -2.71E-04 0.419 0.429 -0.629 0.346 -0.308 -1.05E-03 -0.115

Std. Error 1.107 0.052 0.17 0.003 1.721 0.19 0.581 0.758 0.368 0.04 0.016 0.273 0.257 0.349 0.37 0.359 0.002 0.076

Beta -0.619 0.138 -0.019 -0.031 -0.023 -0.228 -0.048 -0.042 -0.07 -0.051 0 0.035 0.041 -0.042 0.023 -0.022 -0.011 -0.041 6.077 -0.873 -1.349 -1.036 -10.245 -2.242 -1.954 -3.181 -2.259 -0.016 1.537 1.672 -1.8 0.936 -0.856 -0.422 -1.516 0.536 0 0.383 0.178 0.3 0 0.025 0.051 0.001 0.024 0.987 0.125 0.095 0.072 0.349 0.392 0.673 0.13

Table 4. Juvenile Delinquency (Times Last Month) (15-point Scale) (R2 (adj.)=0.037)
Unstandardized Coefficients Std. B Error (Constant) KIDAGE AGE AT TIME OF INTERVIEW 1995 KIDSEX HHINC HH INCOME IN 1000S MOMHISP MOM IS HISPANIC MOMBLACK MOM IS BLACK MOMASIAN MOM IS ASIAN MOMNATAM MOM IS NATIVE AMERICAN MOMOTHER MOM IS OTHER RACE MOMEDUC MOTHERS YEARS OF SCHOOLING MOMAGE SELDOM DOES KID STAY W DAD ONCE OR TWICE? 1=Y, 0 = N SEVERAL DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 =N MONTHLY DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 =N WEEKLY DOES KID SEE DAD MORE THAN WEEKLY? 1=Y, 0 =N OFTEN 16.556 -0.204 -2.395 -0.009 2.747 -0.357 -0.435 1.211 1.861 -0.142 -0.017 0.372 0.583 5.37E02 -0.162 -0.07 2.098 0.099 0.321 0.006 3.003 0.358 1.087 1.406 0.697 0.076 0.031 0.514 0.486 0.66 0.7 0.678 -0.048 -0.166 -0.036 0.021 -0.023 -0.009 0.019 0.06 -0.043 -0.013 0.017 0.03 0.002 -0.006 -0.003 Standardized Coefficients Beta 7.892 -2.068 -7.464 -1.54 0.915 -0.997 -0.4 0.861 2.669 -1.872 -0.552 0.724 1.2 0.081 -0.232 -0.103 0 0.039 0 0.124 0.361 0.319 0.689 0.389 0.008 0.061 0.581 0.469 0.23 0.935 0.816 0.918 t Sig.

26

TALKBIOD TIMES LAST YEAR KID TALKED WITH BIO DAD HOWCLOSE HOW CLOSE KID FEELS TO BIO DAD

4.74E03 -0.584

0.005 0.143

0.028 -0.112

1.014 -4.089

0.311 0

Table 5. Morbidity (Chances of Dying or Being Killed Young) (R2 (adj.) =.030)
Unstandardized Coefficients B Std. Error (Constant) KIDAGE AGE AT TIME OF INTERVIEW 1995 KIDSEX HHINC HH INCOME IN 1000S MOMHISP MOM IS HISPANIC MOMBLACK MOM IS BLACK MOMASIAN MOM IS ASIAN MOMNATAM MOM IS NATIVE AMERICAN MOMOTHER MOM IS OTHER RACE MOMEDUC MOTHERS YEARS OF SCHOOLING MOMAGE SELDOM DOES KID STAY W DAD ONCE OR TWICE? 1=Y, 0 = N SEVERAL DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 = N MONTHLY DOES KID STAY W DAD SEVERAL TIMES? 1=Y, 0 = N WEEKLY DOES KID SEE DAD MORE THAN WEEKLY? 1=Y, 0 = N OFTEN TALKBIOD TIMES LAST YEAR KID TALKED WITH BIO DAD HOWCLOSE HOW CLOSE KID FEELS TO BIO DAD 0.177 2.44E-03 1.15E-04 0 2.19E-02 4.37E-02 1.46E-02 5.28E-02 5.53E-02 -0.007 2.47E-04 1.74E-02 2.36E-02 1.38E-02 -9.09E-03 6.46E-03 -8.44E-05 -6.55E-03 0.046 0.002 0.007 0 0.073 0.008 0.024 0.032 0.015 0.002 0.001 0.011 0.011 0.015 0.016 0.015 0 0.003 0.026 0 -0.035 0.007 0.125 0.013 0.037 0.079 -0.094 0.008 0.036 0.055 0.023 -0.015 0.011 -0.023 -0.057 Standardized Coefficients Beta t 3.833 1.124 0.016 -1.5 0.299 5.533 0.611 1.663 3.565 -4.047 0.361 1.539 2.201 0.948 -0.584 0.432 -0.816 -2.079 Sig. 0 0.261 0.987 0.134 0.765 0 0.541 0.096 0 0 0.718 0.124 0.028 0.343 0.56 0.666 0.414 0.038

Table 6. Descriptive Statistics N 3,786 3,803 3,743 3,784 3,184 3,178 3,757 3,755 3,755 3,788 Minimum 0.00 1.00 18.00 18.00 0.00 0.00 0.00 0.00 0.00 0.00 Maximum 36.00 9.00 61.00 66.00 17.00 17.00 1.00 1.00 1.00 1.00 Mean 10.24 1.89 33.90 36.52 12.87 12.92 0.46 0.36 0.41 0.24 Std. Deviation 6.47 1.04 7.54 8.03 1.99 2.20 0.50 0.48 0.49 0.43

LENGTH OF MARRIAGE CHILDREN IN HOUSEHOLD WIFE'S AGE HUSBAND'S AGE WIFE'S YEARS OF EDUCATION HUSBAND'S YEARS OF EDUCATION WIFE REPRESENTED BY COUNSEL (1=YES) HUSBAND REPRESENTED BY COUNSEL NEITHER SPOUSE REPRESENTED (1YES) BINARY: WAS THERE A DOMESTIC VIOLENCE ALLEGATION? (1-YES)

27

BINARY- WAS THERE A DOMESTIC VIOLENCE PROTECTIVE ORDER ISSUED (1=YES) RESOLVED BY AGREEMENT (1=YES) FEES CHARGED BY COURT INCIDENTS BANKRUPTCY VISITATION OR CHILD SUPPORT MOTION (1=YES)(NOTE: POOLED DATA) CHILD SUPPORT AMOUNT ($) JUDGMENT (USUALLY BACK CHILD SUPPORT, $) HUSBAND PETITIONED (1=YES) WIFE PETITIONED (1=YES) SPOUSES CO-PETITIONED (1=YES) CONTEMPT ACTION (1=YES) RESOLVED BY MEDIATION (1=YES) MEDIATION, NO SETTLEMENT (1=YES) WIFE HAS CUSTODY (1=YES) HUSBAND HAS CUSTODY (1=YES) JOINT CUSTODY (1=YES) PER CAPITA INCOME OF COUNTY (#) UNEMPLOYMENT RATE OF COUNTY (%) MORE D.V. ALLEGATIONS THAN ORDERS (1=YES) INCIDENTS/YEARS 2004-YEAR OF DIVORCE (#/YR) WIFE LIVES OUT OF STATE (1=YES) HUSBAND LIVES OUT OF STATE (1=YES) VALID N (LISTWISE)

3,788 3,780 3,763 3,780 3,768 22863 2,795 3,530 3,806 3,806 3,806 3,765 3,740 3,741 3,806 3,796 3,796 3,806 3,806 3,797 3,792 3,806 3,806 2,058

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 16927.00 2.20 0.00 0.00 0.00 0.00

1.00 1.00 2739.21 283.00 1.00 1.00 6900.00 540000.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 36356.00 14.00 1.00 95.50 1.00 1.00

0.21 0.86 302.10 34.22 0.01 0.0167 357.62 4109.05 0.28 0.57 0.15 0.09 0.08 0.04 0.61 0.09 0.27 26373.34 6.10 0.05 7.66 0.04 0.08

0.41 0.34 177.94 28.86 0.10 0.12801 464.33 23340.27 0.45 0.49 0.35 0.28 0.28 0.20 0.49 0.28 0.44 4245.41 1.95 0.22 7.99 0.20 0.28

Table 7. Joint Custody (Cox & Snell R2=.036) SEPARATION AFTER CUSTODY STATUTE LENGTH OF MARRIAGE SPOUSE ON WELFARE HUSBAND'S NUMBER OF MARRIAGES PROTECTIVE ORDER ISSUED? NUMBER OF CHILDREN IN HOUSEHOLD CONSTANT B 0.256 0.033 -0.454 -0.117 -0.721 -0.099 -0.970 S.E. 0.080 0.006 0.143 0.071 0.114 0.043 0.148 Wald 10.324 27.662 10.043 2.745 40.154 5.471 42.785 df 1 1 1 1 1 1 1 Sig. 0.001 0.000 0.002 0.098 0.000 0.019 0.000 Exp(B) 1.292 1.033 0.635 0.890 0.486 0.905 0.379

Table 8. Child Support Amount (R2 (adj.) =.081) Unstandardized Standardized Coefficients Coefficients VARIABLE separation after effective date of statute (1=yes) B -70.832 Std. Error 21.692 Beta -0.075 t 3.265 Sig. 0.001

28

children in household wife represented by counsel (1=yes) incidents/years 2004-year of divorce (#/yr) either lives out of state (1=yes) wife's age per capita income of county (#) population density (#/mile) Predicted probability of joint custody (from Table 7) (Constant)

51.570 165.756 6.097 36.400 2.846 0.012 -0.132 152.148 -242.748

9.578 19.115 1.209 28.729 1.400 0.003 0.026 136.591 79.459

0.108 0.175 0.107 0.025 0.046 0.103 -0.119 0.027

5.384 8.671 5.042 1.267 2.033 4.334 5.058 1.114 3.055

0.000 0.000 0.000 0.205 0.042 0.000 0.000 0.265 0.002

Table 9. Custody and Visitation Motions Following Divorce (Cox & Snell R2=.010; pooled data) B divorce after effective date of statute year of custody motion neither party represented binary- was there a domestic violence protective order issued (1=yes) bankruptcy wifes age Constant S.E. Wald df Sig. Exp(B)

0.614

0.121

25.799

0.000

1.848

-0.167

0.024

47.923

0.000

0.846

-0.928

0.126

54.136

0.000

0.395

0.805

0.108

55.832

0.000

2.237

0.774 -0.044 330.648

0.349 0.007 48.140

4.906 35.473 47.177

1 1 1

0.027 0.000 0.000

2.168 0.957 ###########

Table 10. Mean Values Before and After the Statute Took Effect
VARIABLE RESOLVED BY MEDIATION (1=YES) JOINT CUSTODY (1=YES) CHILD SUPPORT AMOUNT ($) BINARY: WAS THERE A DOMESTIC VIOLENCE ALLEGATION? (1=YES) BINARY- WAS THERE A DOMESTIC VIOLENCE PROTECTIVE ORDER ISSUED (1=YES) MEDIATION, NO SETTLEMENT (1=YES) Separation Before Statute 0.04 0.24 370.13 0.25 0.24 0.02

Separation After Statute

0.13 0.30 341.03 0.23 0.19 0.06

29

CHILD CUSTODY MOTIONS

0.149

0.183

30