Legislative Analyst's Office, April 13, 1999

The Child Support Enforcement Program From a Fiscal Perspective: How Can Performance Be Improved?

Introduction

In California, the child support enforcement program is administered by county district attorneys under the oversight of the Department of Social Services. During the 1990s, the program has been the subject of considerable criticism for poor performance. In this report, we show how program performance is affected by the state's system for funding the program. Additionally, we present two options for improving this system of funding, which in turn will improve the collections performance of the child support enforcement program.

Background

The child support enforcement program has as its primary purpose the collection of child support payments for custodial parents. The state provides these services to parents receiving public assistance through the CalWORKs program (formerly Aid to Families with Dependent Children) and, on request, to non-CalWORKs parents.

As noted above, the county district attorneys administer the program at the local level and are responsible for determining how much to spend in carrying out this task. About two-thirds of the costs of administering the program are reimbursed by the federal government. In addition, the counties receive incentive payments from the federal and state governments which, in many cases, are sufficient to cover the remaining county costs. In 1997-98, the counties spent $446 million to administer the program, and collected $1.4 billion in child support.

Most of the child support collections made on behalf of families receiving CalWORKs grants are used to reimburse the federal, state, and county governments for their costs of funding these grants. Because the federal and state governments fund most of the costs of CalWORKs grants, they are the primary fiscal beneficiaries of these child support collections.

Program Performance Strongly
Affected by Program Spending

There is significant variation among the 58 counties in program performance. In 1997-98, for example, collections per case for CalWORKs families in the best-performing county was about six times that of the worst-performing county. (For technical reasons related to county case closure practices, we do not have a good measure of performance for the non-CalWORKs component of the program. We note that the fiscal impact of the program on the government is due primarily to collections on CalWORKs cases.)

In reports issued in the early 1990s (see, for example, the Analysis of the 1990-91 Budget Bill, pages 707-710), we showed that there is a strong relationship between the amount of resources committed by the counties in administering the child support enforcement program and the amount of child support collected. Specifically, almost half of the variation among the counties in the average amount collected per case was explained by the variation in "administrative effort" (expenditures per case). In other words, the amount of funds spent by a county in administering its child support program had a significant impact on its eventual success in collecting child support. In addition, we examined a variety of demographic variables (such as county unemployment rates) and found that they had no significant effect on performance, after controlling for administrative effort.

We also found that many of the largest counties in the state had relatively low levels of administrative effort and relatively poor performance in collections. Because these counties have large child support caseloads, their relatively poor performance had a significant effect on the state's overall level of performance.

Figure 1 (see page 4) shows that these findings were still applicable in 1997-98. It shows that the administrative effort per case made by the best-performing counties ($1,253) was, on average, more than twice that of the worst-performing counties ($503).

Fiscal Structure Does Not Encourage
Optimal Program Spending Levels

In a separate report issued in 1992 (The 1992-93 Budget: Perspectives and Issues, pages 153-169), we concluded that the fiscal structure for funding the child support enforcement program in California gave counties an incentive to hold spending down to relatively low levels, even though increased spending was likely to be cost-beneficial from a statewide perspective.

This finding remains applicable to the program as it is currently structured. To illustrate, Figure 2 shows the state and county fiscal impacts of an increase in spending under two scenarios that are representative examples of the fiscal context in which many of the counties operate. The first scenario represents a relatively efficient county which collects an additional $4 in child support for every additional $1 spent in administering the program. The second scenario represents a relatively inefficient county which collects an additional $2 for each additional $1 expended.

The figure shows that after accounting for federal reimbursements for expenditures, CalWORKs grant savings from child support collections, and state and federal incentive payments to the counties, the relatively inefficient county (Scenario 2) experiences a net cost of 5 cents per dollar expended and therefore does not have a fiscal incentive to increase spending. This occurs even though the additional spending would result in a net savings to the state (27 cents) as well as to the state as a whole when combining the state and county impacts (county cost of 5 cents and state savings of 27 cents).

We made similar findings in our 1992 report, and concluded by recommending two options: (1) state assumption of responsibility for administering the program, or (2) establishment of an additional incentive program to encourage counties (particularly the more efficient counties) to increase expenditures. The Legislature subsequently enacted a variation of the fiscal incentive program we proposed (Chapter 92, Statutes of 1992), which is currently known as the State Investment Fund Project. The department estimates that the project is cost effective, but it is a small program (about $3.5 million) because only ten counties have chosen to participate.

In spite of the state and federal incentive payments and the State Investment Funds Project, it appears that the counties continue to under-invest in the program. As we discussed above, one reason is that some counties might find that there is no fiscal benefit to the county in increasing expenditures even though the state would benefit. In our field visits, moreover, state and county administrators indicated that counties are reluctant to increase expenditures in the program once they achieve a "no net county cost" situation--that is, where their revenues from incentive payments and CalWORKs grant savings are sufficient to cover their county costs, net of federal reimbursements. In other words, the counties tend to be "risk-averse." As such, once they reach a no net county cost position, they may choose not to increase expenditures in the program even if they believe there is a reasonably good chance of increasing collections by enough to generate sufficient revenues to cover their additional costs.

What Can Be Done to Increase
Program Spending?

As discussed above, our research leads us to the conclusion that one way to improve program performance--with the prospect of also achieving state savings--is to increase spending for program administration (particularly in the relatively efficient counties). From a fiscal standpoint, program expenditures should be increased in each county to the point where marginal collections equal marginal costs.

We offer two alternatives to move toward "fiscally optimal" investment levels in the child support enforcement program. The first is the same option we advanced in 1992: state administration of the program. In this way, the state would have control over the allocation of program resources. We note, in this respect, that SB 407/1994 (Hughes) was based on this approach, but the bill was not passed by the Legislature. A bill has been introduced in the current legislative session--SB 542 (Burton and Schiff)--which also calls for state administration of the program.

Our second option is to establish a new incentive program; but unlike the existing incentive mechanisms, this program is specifically designed to address the tendency of the counties to be risk-averse with respect to increasing expenditures. Under this option, for any county that increases program expenditures by more than 5 percent (the statewide average increase between 1996-97 and 1997-98), the state would reimburse the county to cover any gap between the county's net revenues (incentive payments and CalWORKs grant savings) and net costs. (Expenditures required for the new statewide automation system would be excluded.) The department would conduct audits of county revenue and expenditure reports and would be responsible for determining the annual fiscal effects.

Referring back to Figure 2, a participating county represented by Scenario 1 would not receive any state reimbursement because the additional spending would result in a net savings to the county. A county represented by Scenario 2, on the other hand, would receive a state reimbursement of 5 cents (the net cost to the county) for each additional dollar expended. Note that in this case the state would still achieve a net savings of 22 cents per dollar expended.

If it is determined that additional program expenditures by a county result in a net fiscal cost to the state as a whole (combined state/county fiscal impact), the county would not be eligible for state reimbursements in subsequent years until the department makes a finding that the county has adopted administrative changes that are likely to lead to an increase in its level of efficiency.

If this new program were to have the desired effect of causing counties to increase their administrative effort, it would lead to net savings to the state in most cases (illustrated by both Scenarios 1 and 2) and potentially to the counties as well (Scenario 1). We note that the program could result in a state reimbursement to a county that would have increased spending by the same amount in the absence of the new incentive program. In such a case, the program would, in effect, result in a net cost to the state and a corresponding gain to the county. The state reimbursement, however, would have to be invested by the county in its child support enforcement program, which could ultimately lead to additional collections that result in savings to the state.

Conclusion

We believe that both options would lead to improvement in the performance of the state's child support enforcement program. State administration probably has greater potential to maximize the net fiscal benefit to the state, but could involve program disruptions such as staffing changes. The new incentive structure, conversely, probably would not result in optimal levels of spending but would be relatively easy to implement because it can be grafted onto the existing administrative structure.