INTRODUCTION

As a condition of eligibility for assistance, the IV-A agency must require each R/A to cooperate (unless good cause for refusing to do so is determined to exist) with the child support enforcement program in:

• Identifying and locating the absent parent of a child for whom aid is claimed;

• Establishing the paternity of a child born out of wedlock for whom aid is claimed;

• Obtaining support payments for a child for whom aid is claimed;

• Obtaining any other payments or property due the R/A or child.

Cooperation includes the following actions that are necessary for the achievement of the objectives specified above:

• Appearing at the local IV-D agency as necessary to provide oral or written information or documentary evidence known to, possessed by, or reasonably obtainable by the R/A;

• Appearing as a witness at judicial or other hearings or proceedings;

• Providing information or attesting to the lack of information under penalty of perjury;

• Forwarding to the IV-D agency any child support payments received from the absent parent after an assignment has been made.

When the IV-D agency notifies the IV-A agency of evidence of failure to cooperate, the IV-A agency must act upon that information to enforce the eligibility requirements. The IV-D agency depends upon IV-A to enforce the cooperation requirement.

This report examines two approaches to handling cases where non-cooperation is a factor. It considers existing procedures that involve both the IV-D and IV-A agencies and a variation where action is pursued solely by the IV-D agency.

One Model Office Project intervention involved comparing the disposition of non-cooperation cases processed using conventional IV-A interventions with cases processed by IV-D staff. During the months of April-November 1996, child support staff in Denver and Mesa Counties kept track of cases cited for non-cooperation. In Denver County, they were referred to the IV-A agency for sanctioning. In Mesa County, IV-D staff assumed total responsibility for sanctioning R/As who refused to cooperate with child support enforcement staff.

Manual case logs were maintained during the study period in both Denver and Mesa Counties. Information was recorded for 54 cases in Denver County and 113 cases in Mesa County. In February 1997, the project evaluators reviewed the automated systems for child support (ACSES) and welfare (COIN) to ascertain the status of the cases on the logs. This allowed us to gauge the incidence of client cooperation and discontinuation. It also allowed us to estimate the level of error associated with each way of handling non-cooperation.

One limitation of the analysis is the lack of random assignment. Each county used a single, different procedure for handling non-cooperation, rather than both counties using both procedures on a random basis. Thus, we cannot rule out the possibility that differences in outcome are due to pre-existing county differences rather than the procedural arrangements under study.

Another weakness of the study is the small number of cases recorded on the logs, particularly in Denver County. Child support technicians in both counties were asked to record all instances of non-cooperation. Compliance was obviously less than perfect and it is unclear whether the cases recorded on the logs are representative of all non-cooperative cases handled by the two child support agencies.

Finally, we lack reliable information on the dates on which various actions were taken pertaining to notification, sanction activity and cooperation. This precluded the possibility of assessing whether noncooperation sanctions were speedier using one procedure rather than another, and/or whether cooperation was stimulated in a more timely manner.

Table 1 summarizes outcomes for the two procedures and counties. Certain case outcomes are comparable across the two counties. Thus, an identical 20 percent of non-cooperative cases are discontinued typically because the client moves, or fails to submit a MSR. Approximately 15 percent of cases in both counties have continued sanctions. In these cases, benefits appear to have been reduced and the applicant has made no attempt to cooperate. Finally, in nearly half of the cases in both counties, the client ultimately cooperated (50% Denver, 44% Mesa) although the incidence of cooperation was somewhat higher in Mesa before referral to the IV-D tech or IV-A agency (29% versus 22%), and higher in Denver after referral to the IV-A agency/IV-D technician (28% versus 15%).

One important difference between the two counties was the error rate. In Mesa County, errors appeared to occur in only 5 percent of the non-cooperation cases. In all of these cases, the COIN screen indicated that a sanction needed to be done, but the R/A's check had not been reduced. In Denver County, the error rate stood at 22 percent. In most of these cases, the sanction was never imposed, even though the IV-D technician had requested one (Case #11, #19, #20, #23, #30, #39, #40). In other cases, the sanction was not imposed in a timely manner. For example, in Case #7, the check hold was not imposed during the period of non-cooperation. As a result, the R/A's check was reduced after she had already cooperated. Finally, in three cases, (#45, #48, #53) the client ultimately cooperated but the sanction was not lifted in a timely manner or at all.

To obtain a subjective view on the effectiveness of different approaches to client non-cooperation, we interviewed administrators and technicians in the Denver and Mesa County agencies dealing with child support. We selected technicians who had been involved with some of the non-cooperation cases listed on the logs.

The interviews suggest that both counties appear to favor transferring the non-cooperation function from the IV-A agency to the IV-D agency. For example, Mesa County doesn't "look forward to sending non-cooperation cases back to IV-A." In addition to keeping the sanction process in IV-D, Mesa supports specialization with a single IV-D technician handling non-cooperation for the entire agency. As the agency director explains:

IV-D should do it. It is a matter of case ownership. Client cooperation is important for IV-D and not as important for IV-A. For IV-A, the sanction process represents extra work that they don't like.

Denver County child support technicians are also frustrated by the lack of control they have over the sanction process. They complain that sanctions sometimes do not get done even after they have been requested several times. When a IV-A worker fails to follow through, the child support worker is "left hanging." This undermines their credibility with clients and can slow down case progress. As one technician explains:

I am tough with my clients. I tell them what they need to do and I don't give them extra chances. I tell them that they will be sanctioned if they don't cooperate. So I need to back up my threats with reality.

While some IV-A workers are credited with being "great," others are faulted for having to be "chased" to impose a sanction. Child support workers wonder whether IV-A workers simply do not understand "why we need it and why it is important." In order to be certain that the sanction is imposed when it is needed, this worker would like IV-D to take over the function or to have it generated by the system. As it stands, the non-cooperation process is totally manual with IV-D workers needing to do handwritten notices to IV-A workers to initiate a sanction.

One drawback to transferring the non-cooperation function to the child support agency is the unfamiliarity of most child support workers with COIN. The IV-A agency may be reluctant to make the COIN system accessible to untrained IV-D workers. Some IV-D workers feel that COIN is a fairly easy system to master and that they could handle the non-cooperation cases in their own caseloads. Alternatively, the non-cooperation function could be specialized with a single or limited number of IV-D workers handling the job for the entire agency.

Specialization offers several advantages. If all cases of non-cooperation are handled by a single technician, the training requirements are minimized as is the potential for error. Routing all cases with non-cooperation factors to a single specialist can also promote uniformity of case treatment. Finally, the single specialist may help improve relationships between the IV-A and IV-D agencies by acting as a "go-between."

In Mesa County, for example, the non-cooperation function is handled by a single child support worker who had previously worked in the IV-A agency. In addition to being familiar with both the COIN and ACSES systems, this individual had helpful relationships with workers in both agencies. Finally, assignment of all non-cooperation duties to a single technician is perceived to reduce the chance of neglecting to impose a needed sanction or lifting a sanction once cooperation is achieved. As this worker explains:

I enjoy doing non-cooperation. I monitor my non-coop cases regularly. Sometimes the sanctions don't take, so I always check my screens to make sure that they are imposed. If she cooperates, I make sure the sanction goes off. I get the information from IV-A and IV-D and I can kind of keep both sides informed. You can't always go by the screens and know what is really going on in a case.

This technician feels that the traditional division of labor in the non-cooperation process is ineffective for a variety of reasons, not the least of which is the IV-A technicians' discomfort with imposing a sanction without fully understanding the case details. When she worked as a IV-A technician, she recalls feeling as though IV-D technicians should be the ones to talk to clients about the importance of cooperation and was unwilling to engage in this dialogue with non-cooperative clients.

Specialization of the non-cooperation function in Denver County presents some unique challenges. Staff doubt whether a technician could handle non-cooperation duties for the whole agency in addition to their normal caseload. One suggestion is to lodge the specialized, non-cooperation function with intake workers who might have fewer time pressures than technicians on regular establishment and enforcement teams.

If the non-cooperation function cannot be transferred from IV-A to the IV-D agency, technicians would like to have it specialized in the IV-A agency. They would prefer dealing with the lead workers on the IV-A units rather than all the workers. Lead workers have generally been extremely responsive. The hope is that by concentrating responsibility for non-cooperation with a limited number of workers, communication and performance will improve.

As part of the Model Office Project, Denver and Mesa Counties processed cases with non-cooperation factors using two different approaches. In Denver, the IV-D technician requested that the IV-A worker impose a sanction when a client was non-cooperative and failed to take necessary remedial steps. In Mesa, the IV-D technician made the same request to a specialized IV-D technician. Records were kept of all cases handled using these two different approaches.

A comparison of outcomes suggests that the error rate is dramatically higher when the IV-D technician must rely on the IV-A worker to perform the sanction. In Denver County, the error rate stood at 22 percent as compared with 5 percent in Mesa where the IV-D specialist handled all non-cooperation matters.

Interviews with Denver and Mesa County technicians confirm that child support workers would like to retain more control over the sanction process. Denver County technicians want to handle noncooperation in their own caseloads. In Mesa County, specialization of the non-cooperation function in the child support agency is regarded as one way to limit worker access to COIN, reduce training requirements, promote uniformity of case treatment, minimize error and build relationships between the IV-A and IV-D agencies.

If the IV-D agency cannot acquire the non-cooperation function, technicians suggest that these duties be performed by lead workers in the IV-A units rather than by all workers. This type of specialization might serve to promote uniformity of case treatment and sharpen accountability. Finally, it is hoped that this change would improve agency response when a client fails to cooperate.