In 1990, the federal government imposed a requirement on states to use rigid formulae (known as “guidelines”) to determine the amount of child support non-custodial parents are ordered to pay. If states did not comply, they faced loss of federal funding. Since then, parents and analysts have been aghast by the arbitrariness of award amounts. Underlying problems in developing a better formula have been solved. What remains is the politics involved in putting the knowledge to use.

The federal requirement developed in what was at least in part a backlash against early work on developing child support formulae. Most notably, Maurice Franks authored a formula that was being recognized by state courts as coming close to matching established policy. Franks' work was in fact an effort to model existing policy and he did not go so far as to claim any more fundamental or scientific basis to his formula. A similar effort was carried by Judge Melson became the model for the the first version of Delaware's child support guidelines. Similar efforts were being made throughout the country by individuals and local bar associations.

Groups representing divorced and never-married mothers saw the emerging change in process as an easily obfuscated back-door route to policy change. They were soon joined by entrepreneurs who sensed an opportunity for profit connected to greater federal involvement in marriage and divorce. When the requirement went into effect in the guise of “child support enforcement” it came with recommendations from federal regulators to increase awards to two and a half times what they had been under established state law and provided financial incentives to states for the increase.

Advocates of the new formulae implied a scientific essence in their work by saying it is based on “economic studies.” Their studies consisted of manipulating statistical data on family spending and presenting estimates of the cost of raising children. They constructed simple equations to divide the cost between parents. It is more widely recognized now that their equations are too simple to make sense in a wide range of circumstances and that their cost of raising children estimates are mere fabrications. In other words, their formulae are arbitrary manipulations of child support policy. (journal article) .

Intelligent Systems Research Corporation (ISR) began work on the guideline design problem in the late 1980s. After reviewing the “economic studies” approach, attention turned to the approach previously taken by state courts, beginning with review of work by Franks, Melson, and others.

It was thought at that time that courts would eventually recognize the arbitrary nature of existing guidelines and insist on constitutional grounds on a return to the previously established, time-tested model. (MND article) ISR planned to develop general mathematics that could be adapted to the nuances in child support and welfare policy in individual states.

The prediction that courts would begin correcting the problem did not hold true however. In 1993, the U.S. 9th Circuit Court of Appeals issued a ruling in P.O.P.S. v. GARDNER that reclassified family policy from a private issue to “social policy,” in effect socializing child support. This effectively eliminated the application of constitutional principles.

But ISR was on the verge of a significant discovery. Traditional child support policy that had developed over generations through repeated analysis in (at least) the shadow of the constitution was more than just a good idea. Evidence emerged that there may be only one correct theory of child support determination. Traditional due-process had paralleled scientific process to some degree in the search for it.

The work itself transferred from ISR to a private project called Project for the Improvement of Child Support Litigation Technology (PICSLT). In 1994, PICSLT presented a new formula focusing first on the logic of child support decisions rather than cost estimates. It might be said that the key to the formula is that “child support” is taken to be just that – child support. There is a natural logic to the decisions so long as they are unbiased.

In a more recent discussion paper, PICSLT presents (in a little more than 14,000 words) a review of traditional state court precedents in child support cases, extracts the main principles, and reexamines them in a more generic way. When properly understood, the principles that had been stated in traditional law and repeatedly examined by courts are scientifically valid. The paper goes on to show that deviation from the principles yields answers that quite recognizably incorrect.

Roger F. Gay