January 2012

Virginia seems likely to increase its child-support obligations on non-custodial parents, based on flawed methods of calculating child-rearing costs that overstate child-rearing costs. As I explain below, the result is likely to be excessive child support obligations for most non-custodial parents in the state. (Virginia’s child-support guidelines already assume that parents spend far more on their children than couples like I and my wife actually spend. To spend as much as Virginia’s guidelines require non-custodial parents to pay in child support, we would have to do something really expensive, like feeding our daughter steak at every meal, and buying her lots of designer clothing.)

State child-support agencies may have a built-in incentive to push for increases in child-support obligations, even if they are unwarranted, since when child-support obligations increase, it becomes more difficult for non-custodial parents to pay those obligations, which creates a need for more funding for the state child-support agency to collect those obligations. (It also creates a need for more funding for state correctional systems, since failure to pay child support can result in incarceration, under an exception to the usual rule that imprisonment for debt violates the Thirteenth Amendment, resulting in thousands of people being jailed who are unwilling or unable to pay their child support. (See United States v. Ballek (1999), which carved out an exception to the Thirteenth Amendment for child support, and Turner v. Rogers (2011), which discusses states’ practice of jailing non-custodial parents who are behind on their child support payments, including those who might be unable to pay. In Mahoney v. Mahoney (2000), the Virginia Court of Appeals refused to hear the appeal of a father jailed for non-payment of child support, due to his failure to put up an appeal bond. The requirement to put up an appeal bond can create a Catch-22 situation for parents, because if they are too poor to pay their child support obligation, then they may also be too poor to pay for an appeal bond).)

The Virginia Department of Social Services, which contains Virginia’s child support agency, has hired Jane Venohr of the Center for Policy Research to draft the state’s new child-support schedule (after no competitive bidding). She has a history of issuing reports on state child support guidelines that have been cited to push through dramatic increases in child support obligations, based on the methods she uses to estimate child-rearing costs. This no doubt makes her popular among state and federal child-support officials, for whom more is better when it comes to child-support obligations.

In the past, Ms. Venohr has advocated using the generous Rothbarth method as a floor for calculating child-support costs, arguing that child support levels should fall between the amount generated by two methods — the Rothbarth method and the still-more generous Engel method — that, as I will explain below, both overestimate child-rearing costs.

(Economist R. Mark Rogers explains why these methods overestimate child-rearing costs here. Mr. Rogers was a long-time economic forecaster for the Federal Reserve Bank of Atlanta.)

Ms. Venohr has written that “the Engel methodology overstates actual child-rearing expenditures and the Rothbarth methodology understates actual child-rearing expenditures.”

But both of these methods lead to excessive costs, not just the Engel method. Both methods are very crude ways of estimating child-rearing costs that economists use only out of perceived necessity, given the supposed unavailability of alternative methods of calculating with precision what the average person actually spends on their kids. (I hope to discuss better alternatives in future commentaries.)

Australia’s child support agency advocates using the Rothbarth method as the least-bad method, but it nevertheless admits that both methods are very flawed methods of estimating child-rearing costs, in a commentary entitled “Costs of children and equivalence scales.” As it notes, the Rothbarth method compares “expenditures on goods that can be attributed” to adults, like liquor and tobacco, and compares how much less parents spend on these things than childless couples. “Expenditure on adult goods (e.g. alcohol, tobacco and adult clothing) should decline when a child is added to the family as resources are diverted from adult goods to meeting the needs of the child. The Rothbarth approach imputes the same welfare level to households that have the same level of consumption of adult goods. The Rothbarth method defines the costs of children as the reduction in income which would lead to the same reduction in expenditure on adult goods that the addition of a child to a family generates.”

As the agency notes, “A number of criticisms have been made of the Rothbarth method. Perhaps the most telling is that although children do not consume adult goods, their presence may alter their parents’ tastes for adult goods. Similarly, the presence of a child is likely to change the way the parents spend their leisure time and ‘surplus’ cash. This makes it very difficult, if not impossible, to find adult goods for which family consumption is not directly affected by the presence of children. In practice, tobacco and alcohol are often used as adult only goods and it seems rather strange to equate welfare with consumption of these goods. Both the Engel and Rothbarth methods ignore the impact of the addition of a child to a household’s preference between items.”

As economist Rogers puts it, “If parents want to share household shared goods with children, then the Rothbarth methodology overstates child costs,” since a tendency to share –rather than impoverishment — may explain a fall in consumption of adults-only goods after the arrival of children. If “after having children, the parents have a preference to spend more time with shared goods” — like me and my daughter watching our living-room TV together, even though I didn’t watch much TV there before the birth of my daughter — “then the Rothbarth methodology overestimates child costs.”

The Rothbarth method also overstates child-rearing costs because it assumes that reductions in consumption of adult goods are the result of child-rearing costs, rather than increases in the effective price of adult-only goods due to the presence of the children that in fact have little effect on the household’s overall living standard. “For example,” notes Australia’s agency, “the birth of a child will increase the price of outside entertainment for a couple if babysitting services need to be paid for,” reducing the amount of outside entertainment they purchase even if the child costs very little to raise.

I and my wife spend very little on our daughter compared to the amount households with our income level are ordered to pay under Virginia’s existing child support guidelines, which assume that the average household spends lots of money on their kids. Yet our consumption of adult-only goods like tobacco, alcohol, and forms of entertainment aimed at adults has crashed since the birth of our daughter.

We no longer go to adult-oriented comedy clubs, like the Improv, because the material wouldn’t appeal to our daughter, and she is rather young for a baby sitter. My wife now smokes less, to reduce our daughter’s exposure to second-hand smoke. I now consume less wine, because when I am caring for my daughter, I don’t want to risk a spilled wine glass, don’t want to lower my energy level (even if it would be relaxing), and don’t want to tempt my daughter to drink my wine. Owing to scarce time left over after caring for our daughter, my wife has bought far less adult clothing since the birth of our daughter.

Using the Rothbarth methodology, which assumes that a reduction in spending on adult-only goods is the result of child-rearing costs, it would seem like we spend much of our income on our daughter. But nothing could be further from the truth. In reality, we actually spend only a few percent of our household income on our daughter, and the costs attributable to her have never approached half of what Virginia’s child support schedule assumes households with our income typically spend on their children.

As Australia’s child-support agency notes, “The logic underlying the Rothbarth method is that children brings needs but no resources to a family, and those needs can be met only by making cuts elsewhere in the budget.” This is itself a flawed assumption that can lead to overestimates of the cost of raising a child, since in developed countries, “families with children receive government child related income supplements to assist with the costs of bearing and raising children so that in reality children bring additional resources to a family.” For example, in the U.S., households receive a refundable child tax credit (which lower income-households can receive from the government even if they pay no income tax), a personal exemption of $3650 per child from taxable income, and tax credits from the federal government to cover part of the cost of providing daycare for the child (as well as a tax exemption from the State of Virginia for the same expenses).

For many lower-income households, tax benefits like the refundable tax credit amount to most of the cost of raising a child. Upon a divorce, tax credits and exemptions are typically claimed by the custodial parent, to the exclusion of the non-custodial parent. (In Virginia, unlike most states, judges have no authority to order the custodial parent to waive the child tax exemption or credit so that it may be claimed by the non-custodial parent, even if the non-custodial parent provides most of the financial support for the child. See Floyd v. Floyd, 436 S.E.2d 451, 463 (Va. 1997); Pearlene Anklesaria, Child-Related Tax Breaks for Divorced Parents, 22 J. Am. Acad. of Matrim. L. 425, 426-27 (2009) (contrasting Virginia law with the law of most states). The custodial parent is also more able than the non-custodial parent to claim the Earned Income Tax Credit.

These tax benefits are not reflected in Virginia’s child support schedule, although they should be. Instead, Virginia law leaves courts with the discretion to consider taxes as a “deviation factor” in reducing or increasing child support obligations (except for day-care-related tax credits and exemptions, which courts must consider if the other parent actually shows the amount of the resulting “tax savings”). But in practice, these tax benefits are not usually addressed by Virginia courts in child-support rulings. The courts also refuse to allocate the tax benefits of raising children to a non-custodial parent even when that parent pays for most of the child’s living expenses. Even if most litigants knew about these tax issues — and were inclined to litigate them — they could be hampered by their lawyers’ lack of knowledge. Calculating the value of these tax benefits to custodial parents requires math skills, which lawyers often lack. A classmate of mine at Harvard Law School, who went on to become the General Counsel of a Fortune 500 company, told tax-law professor Ed Warren that 10 percent of 100 is 20! Due to mathematical illiteracy, lawyers sometimes mess up the calculation of their clients’ sentences under the very simple formulas provided in state and federal sentencing guidelines, as Slate noted in an October 22, 2009 article by Ray Fisman, entitled “Errors in Judgment.”

Thus, although these tax benefits affect most households, they will not be taken into account unless they are built into the child-support schedule itself, by reducing child-support obligations to reflect the amount already received by the custodial parent in tax benefits. Thus, the child-support schedule should take into account the fact that custodial parents can typically claim a refundable $1,000 tax credit for each child on line 51 of their tax return (see IRS Form 1040), even if they pay no income taxes, and can claim a tax exemption of $3650 per child on line 42 of their tax return.)

(The IRS Form 1040 can be found here. The $3,650 tax exemption per child is listed right on line 42 (“Exemptions. Multiply $3,650 by the number on line 6d.”) The additional tax credit of  $1,000 per child claimable on line 51 is made clear on pg. 43 of the instructions for the IRS Form 1040 form. See 1040 Instructions, available here. In addition, a federal tax credit of up to around $1,000 per child is available for day-care, on line 48 of the Form 1040, called the “Credit for child and dependent care expenses.” (The credit can be computed on IRS form 2441, which can be found here.) The custodial parent’s daycare expenses can also be deducted on the Virginia tax return, see “Child and Dependent Care Expenses Deduction” available here. Yet non-custodial parents are ordered to pay day-care expenses anyway, on top of the amount contained in the child-support schedule, as the Virginia Court of Appeals decision in Herring v. Herring (2000) illustrates.  In that case, $667 per month in daycare was added on top of the $675 per month from the child-support schedule.  See Va. Code § 20-108.2(F). Custodial parents with incomes below $35,000 a year can also claim a refundable earned income tax credit for qualifying “children” who “lived with” them on line 64a of the Form 1040.  See 1040 Instructions at pp. 45-46.

Ms. Venohr was correct to admit that the even more generous Engel method of calculating child-rearing costs leads to excessive estimates.  (Economist Mark Rogers agrees on that point, in his June 28, 2011 commentary, “Documenting that Both Engel and Rothbarth Versions of Income Shares Cost Tables Overestimate Child Costs.”  That commentary, which I quoted from above, is found at www.guidelineeconomics.com/files/GA_cscomm/engel_rothbarth_overestimate.pdf)

As the Australian agency notes, the Engel method overstates child-rearing expenses, since it bases its estimate of child-rearing costs on how much of a household’s spending on things like food is attributable to the kids. This inflates the estimate of child costs, since kids account for a bigger percentage of spending on food than on many other expenses, since two people can’t eat the same piece of food, but they can share use of other family assets, like sharing the same living room, kitchen, car, etc. As the agency notes, the Engel method “uses the share of the family budget devoted to food as an indicator of living standards,” assuming that “all that is needed to calculate the cost of a child is to calculate how much must be added to the budget to restore the family’s food share to its original value.” But “the Engel and iso-prop methods are only valid if the assumption that the proportion of the budget spent on food (or other necessities) correctly indicate family welfare. A major limitation of the Engel method is that since a child consumes mostly food and clothing, providing an income which will allow the share of the family budget spent on food to return to the pre-child level will overcompensate the family for the addition of a child,” systematically overstating the cost of a child.

In conclusion, I should note that I am not divorced, do not owe child support, and do not have any children out of wedlock. Thus, I have no immediate stake in this issue. I have, however, read hundreds (if not thousands) of court decisions dealing with child support, and the child support guidelines of most states. I also have an economics degree (including some study of economic modeling and a short stint at the Bureau of Labor Statistics) and a law degree.