Summary of shared parenting research from Linda Nielsen

Professor Linda Nielsen of Wake Forest University presented at the International Conference on Shared Parenting 2017. We interviewed Nielsen for a chapter in Real World Divorce:

What does Nielsen think of the winner/loser custody system that prevails in most U.S. states? “A lot of social scientists say that a court cannot possibly pull together enough custody evaluators and psychology experts to accurately predict what is going to be the best parenting plan for each child in a particular family ” responded Nielsen. “The premise that custody evaluators can always give an objective recommendation is flawed. It is not like a driving test or a math test. There may be no standard set of credentials for custody evaluators. There is not necessarily consistency from one evaluator to another and many of the measures used in these evaluations were not designed for that purpose.. A psychologist can’t walk into an intact family, do an assessment and determine which parent is better for which child at which age in that family – or who will be the better parent four years from now. So why bring that difficult task into family court?”

Nielsen says that a deeper problem with courts picking the “better parent” at the time of divorce may be that the judge is answering the wrong question. “It doesn’t matter who is a better parent at the time of the divorce,” says Nielsen. “I ask students [in a Wake Forest University Department of Psychology course] ‘Was your mother or father the better parent when you were 6, 10, 16 years old? Now answer the same question for your brother or sister. The answer is different at each age and, with siblings, depending on the personality of the siblings and the parents. The importance or effectiveness of each parent will go up and down as the child ages, which is one reason that children who are in shared parenting arrangements do better than children who spend less than 35 percent of their time with one parent.”

In other words, if Nielsen’s research-informed perspective is correct, states are collectively spending billions of dollars of tax money (to pay for judges and courthouses) and consumers are spending tens of billions of dollars (on litigators) on an exercise that, in most cases, should not be undertaken.

What about the fact that some divorce lawsuit defendants turn out to be “Disney Dads” post-divorce? Doesn’t that confirm the wisdom of the judges who assigned them a secondary role? “Research shows that the type of parenting you can do depends on what activities and how much time you get to share with the child,” says Professor Nielsen. “The best kind of parenting is called ‘authoritative parenting’, as distinct from ‘permissive parenting’, which is the worst. An authoritative parent sets rules and talks to children about important things. He is a child’s parent, not the child’s uncle. For this to be possible the children must spend ample time with the father and have a full range of activities with him, including ample time during the school week. When you cut the parenting time down to every other weekend there’s not an opportunity to be an authoritative parent. It is not that the dad is a different person. He’s the same person with the same parenting skills but in a restricted situation.”

She displayed more of her academic side at the conference, talking about a meta-analysis of 52 studies of shared parenting. She started by looking at all published studies in the English language, which used definitions of “shared” parenting ranging from a 35/65 to a 50/50 schedule. Generally these studies looked at shared parenting versus every-other-weekend-with-the-loser parenting. Nielsen first gave the raw results:

  • 30 studies found children in shared parenting did better on all metrics; 12 were equal or better on all metrics; 6 were better on most, worse on 1; 4 studies showed no difference

Nielsen explains that people tend to dismiss shared parenting studies by pointing out that children in shared parenting tend to have wealthier parents, if only because when the parents have a low income there are only sufficient resources for one household physically large enough to hold the kids (e.g., one parent will have a house, perhaps provided by taxpayers, while the other parent will have a studio apartment). Nielsen explained that the skepticism might be motivated by a false assumption. For children in intact families who aren’t in poverty there are “not strong links between family income and children’s emotional, behavioral, and psychological well-being. In fact, richer kids may do worse.” (see Are rich kids better off overall?)

Nielsen looked the subset of the 52 studies where income data were available. These showed similar benefits to children from shared parenting, with 17 studies finding that children did better on all metrics.

American family courts in a lot of states will deny a defendant’s request for shared parenting if a plaintiff seeking sole/primary custody asserts that there is “conflict” between the parties (see the Massachusetts chapter for some examples). The winner/loser schedule almost always results in more parent-to-parent contact than a week-on/week-off shared parenting schedule, but courts simply posit that there will be less fighting if one parent is reduced to loser/babysitter.

Nielsen looked at 6 studies where the “level of agreement” between the parents was recorded. It turns out that 50-80 percent of separated parents generally fall into the category of “do not agree.” For those “do not agree” parents, shared parenting still resulted in equal or better outcomes for the children.

What about fomenting additional conflict between parents? 15 studies showed the parents of children in shared parenting had an equal amount of conflict compared to children in sole custody, 2 showed less conflict, 1 showed more conflict from shared parenting, and 2 showed mixed results.

Nielsen’s overall conclusion: Adults participating in a shared parenting arrangement did not have higher incomes, less conflict, better cooperation, or better parenting skills.

From my point of view, one big weakness in the American studies is the failure to distinguish between 35/65 and 50/50 parenting schedules. The 35/65 threshold is used for historical reasons (some early studies used it) and, like most things that the research psychologists do, it considers practical issues of litigation and cashflow to be irrelevant to human behavior and emotions. From our Rationale chapter:

In November 2014 we interviewed Margaret Bennett, a prominent divorce litigator in the Chicago area and a member of legislative committees redrafting the core Illinois family law statutes and creating a new child support formula. We asked her what she thought of the existing Illinois winner-take-all system in which one parent is designated as primary and receives potentially unlimited amounts of child support. Did that give parents a cash incentive to fight? The answer was “no,” a typical perspective among attorneys involved in making laws and guidelines in the winner-take-all states.

We gave Bennett a concrete example of two doctors were splitting up, each with an after-tax income of $200,000. If they had two children the parent who won custody would get $56,000 per year. That would work out to roughly $1 million over the 18-year period of child support eligibility in Illinois. If the parents had a 60/40 time split, the parent with whom the children spent 60 percent of the time would be $2 million wealthier than the parent who took care of the kids 40 percent time. Bennett said that “upper income parents spend a lot on their kids” and that this $56,000 per year annual payment would not cover what the victorious parent was actually spending. Why did parents spend so much on legal fees fighting over custody if not because of the $2 million at stake? Bennett responded that it was primarily because “it is a big status symbol to be the residential custodian.”

When the two litigants have similar incomes, an increasingly common situation in family court lawsuits, there is an enormous practical difference between a 35/65 schedule outcome and a 50/50 outcome. At 50/50 the parties will end up with equal spending power. At 35/65, one party will be paying the other and therefore, depending on the state and the number of children, the winner may have 2X the after-tax spending power of the loser. The dynamics of the co-parenting relationship, particularly when stepparents are involved, may be very different when one household is living off the other household’s income compared to when the two parents are financially independent of each other.

Thus I think that the European studies provide better guidance to the psychology per se. They tend to use a strict 50/50 definition of “shared” and the child support formulae in Europe don’t result in large spending power differences even if one parent does emerge as “primary.”

What did Nielsen think about putting presumptions of 50/50 shared parenting into the law? She was for it: “Children with cancer should not be vaccinated, but we still have a presumption of vaccination. It is the same with shared parenting. There are rare exceptions where it wouldn’t be in a child’s best interest.” Professor Kari Adamsons responded with “We have a presumption of innocent, but that doesn’t mean nobody goes to jail.”