COLAs are generated at the state level by the state’s computer system, PRISM.^[14] Work at the county level is minimal. PRISM first selects and processes eligible cases. It then calculates the new order amount and sends notices to the custodial parents, non-custodial parents, and employers (if child support is paid through income withholding).

Not all cases are selected for COLA in a given year. Cases not selected include those that were COLA adjusted in the previous year, cases with original orders or orders recently adjusted through the formal modification process that are less than two years old, and cases with an order that did not originate in the state of Minnesota. Cases where domestic violence is an issue also do not receive automatic COLAs. Exhibit II.2 depicts the COLA process in Minnesota.

Since 1983, court orders have included language stating that COLAs can be applied every two years to child support awards. For example, if the award was established in 1993, COLAs would occur in 1995, 1997, and every odd-numbered year thereafter. The original order indicates which cost-of-living index will be used for the adjustment.^[15] There are four Consumer Price Indexes used for calculating the COLA:

The index used may vary based on the residential location and occupation of the non-custodial parent as determined by the court at the time the court order is established.

COLAs are automatic for all IV-D cases unless the order specifies otherwise.^[17] There is no distinction between TANF and non-TANF cases.

The COLA process for IV-D cases is as follows:

When two years have elapsed since the order’s creation or the most recent COLA, the state IV-D office mails notices of the upcoming COLA to the custodial and non-custodial parents. Notices are generated and mailed by PRISM during the third week of March. (Copies of the notices are contained in Appendix A .) The effective date for all COLAs is May 1.

The notice for the non-custodial parent lists the following information:

The notice for custodial parents lists the following information:

In addition to informing the custodial and non-custodial parents of the upcoming COLA, the computer generates a notice in mid-April informing the employer (in cases of income withholding) of the new support amount, effective May 1. The notice is sent prior to the effective date so that employers will be able to adjust withholdings for the first paycheck in May.

Exhibit II.3 describes the COLA calculation for an order with an obligation amount of $200 per month. In this example, the CPI-U has increased by 3.6% during the two-year period, which translates to a $7.00 adjustment to the order.

Appeals are referred to as “motions to stay.” The process is the same regardless of whether the IV-D case is TANF or not. The state sends COLA notices to the custodial and non-custodial parents in the latter part of March. If non-custodial parents dispute the COLA, they must file a motion to stay no later than April 30 of that year. If there is no motion filed, the COLA will be effective May 1.

Documents necessary for filing a Motion to Stay COLA can be obtained from the county child support agency or the county court administrator’s office (see Appendix A for copies of forms). The non-custodial parent must complete the motion papers, attach tax information for the previous three years, and sign the Notice of Motion and supporting affidavit before a notary public. The non-custodial parent must then file the motion with the court administrator to obtain a hearing date and serve the motion upon the custodial parent and the IV-D agency. Papers must be filed and served at least 20 days prior to the court date. There is a $25 filing fee ($120 if the non-custodial parent has not filed legal papers with the court in the past).

Other parties do not request motions to stay COLAs. There is no set procedure for custodial parents to appeal. A custodial parent who suspected the non-custodial parent’s income increased by more than the cost of living would have to file a modification petition in court. Moreover, the county IV-D agencies do not appeal COLAs on behalf of their cases. Wage information is confidential and cannot be disclosed outside of court. Thus, IV-D agencies cannot alert custodial parents if the incomes of non-custodial parents outpace inflation.

Once a motion to stay has been filed, the PRISM system sends a notice to the employer to prevent an adjustment to the income withholding. The county child support officer collects matching wage information for the county attorney, who represents the IV-D office in court. The county attorney represents all IV-D cases, regardless of whether they are TANF cases or not.

Court magistrates hear the motions to stay. The hearings are short—generally about 15 minutes. With a motion to stay, courts have very limited review. Testimony is limited to the non-custodial parent’s income. The non-custodial parent must bring three years of tax data to court; failure to do so will result in the COLA going into effect. If the tax information shows that wage growth was greater than the CPI, the stay is denied. If the wage growth is less than the CPI, the magistrate has some discretion.^[19] For example, if the non-custodial parent’s income increased 3% while the COLA was 6%, the magistrate could adjust the award by 3%. If the non-custodial parent’s income declined, the appeal process would not result in a downward adjustment; the magistrate would simply not apply the COLA.

This judicial process was adopted in 1999 after the state Supreme Court ruled that the previous appeals process, known as the administrative process, was unconstitutional. Under the administrative process, non-custodial parents who wanted to contest their COLAs called the IV-D child support officer. The child support officer then collected the relevant information and forwarded it to an administrative law judge, who scheduled a hearing. Consequently, when compared to the judicial process, the time involved from the perspective of non-custodial parents was minimal.

Changes to the computer system have been common. The computer system, and its central processing of notices and adjustments, was vital to the success of the COLA implementation. Since the PRISM system’s implementation in 1997, there have been fewer complaints from county staff about the time burdens associated with COLA. The system generates the following reports:

Glitches in the system remain. The computer system can be rigid, and it is difficult for a child support officer to change an error. Although staff manually correct the order amount in the system, other fields need correcting to prevent the COLA from occurring during the wrong cycle in future years. The COLA effective date can be altered, which will put the COLA on the right cycle.

To address some of these concerns, the state continues to develop educational materials for staff. In 2000, the state issued two detailed policy memos that spelled out statutory and computer procedures. The state also developed a COLA policy video that was distributed to counties via satellite. Among other topics, it explained what Department of Employment Security information was required by the courts in the appeals process. It also illustrated how to deal with complicated cases (e.g., multiple orders, emancipated children, interstate cases). Counties also provide their own training, although not much discretion is left to the counties in terms of procedures.

The state Child Support Enforcement Division also operates a help desk daily from 8 a.m. to 4:30 p.m. County staff can call or e-mail questions. In addition, there are daily, one-hour meetings on policy and technical issues. Staff can call in and participate.

Staff we interviewed stated that the COLA required minimal work, especially when viewed over the course of a year. COLA-related work peaks during the January to March case selection period and then declines considerably. At the state level, the lead computer analyst spends about 50% of her time on COLA issues between mid-December and mid-April. One other programmer spends 20% to 25% of time during same period. A COLA-testing staff person also works on the COLA, although the amount of time decreases each year. During the other months, staff spends minimal time on COLA. In the policy office, the COLA program advisor works full-time on the issue from January through March, half time in April, and about 5% of the time during the other months.

The COLA also consumes little staff time at the county level. In Ramsey County (St. Paul), clean-up of the COLA reports is completed in a matter of days. The 40 child support enforcement officers each spend about 12 hours each year to review and clean up their COLA-eligible cases. Each has about 500 cases, so roughly 250 are eligible in any given year.

The new judicially based process, which was mandated by the Minnesota Supreme Court in 1999, has limited the role IV-D caseworkers play in COLA appeals. For example, in Hennepin County (Minneapolis), IV-D staff spent about two hours per case under the pre-1999 administrative process because they had to collect financial information and serve papers to all parties. Under the new process, staff spends about 20 minutes on each appealed case. In addition, the number of contested COLAs declined county-wide, from about 400 to less than 100. As discussed in more detail below, much of the paperwork associated with appeals shifted from IV-D to the non-custodial parents, which may be a factor in the decreasing number of appeals.

The time spent on COLA-related business also pales in comparison to what was spent conducting reviews and adjustments. Child support officers in Ramsey County noted that the process for one review and adjustment case can take up to six months to complete, and no less than three months. It is very labor intensive because the IV-D office has to collect all the necessary information (income, medical support, child care expenses). The agency then sends statements to both parents requesting income, expenses, and insurance data. IV-D applies the data collected to the state guidelines. The county IV-D office sends the new order to both parents, who have one month to dispute the findings. If there is a dispute, a hearing is set. Even though four staffers work full time on reviews and adjustments, there is still a large backlog.

Finally, COLA-related business has a limited effect on the court calendar. According to the court administrator in Washington County, COLAs affect the court’s workload in May—and, to a lesser extent, in April and June. Over the course of the year, however, the COLA policy decreases the court’s workload by reducing income-related modifications principally to those with large changes in income. COLA hearings are straightforward. The magistrate reviews three years of income tax returns. The hearing usually lasts about 15 minutes. By contrast, the standard modification hearing may last an hour or more. Moreover, the magistrate generally spends several hours reviewing all of the documents (both parents’ incomes, child care, medical) prior to the standard modifications hearing. The COLA is not meant to be a substitute for a traditional review and adjustment—parents can and still do request it. For example, in 1998, the state modified 3,825 cases through the traditional review process.

Each year, Minnesota evaluates the COLA’s impacts on obligations and collections. The state’s data indicate that in 2000, 29% of the caseload with orders was COLA eligible (i.e., 48,169 cases selected for COLA out of 168,550 cases with orders) (see Exhibit II.4 ). Cases not selected for COLA include those that were COLA adjusted in the previous year, those that were recently adjusted through the formal modification process, those that are less than two years old, and those with an order that did not originate in the state of Minnesota. Other orders not selected have specific terms with regard to COLA. The data indicate that the share of TANF cases with orders that were COLA eligible increased from 35% in 1998 in 50% in 2000. The trend suggests that the TANF caseload is now composed of longer-term cases that are more likely to satisfy the two-year review criterion. Meanwhile, the share of non-TANF cases with orders deemed COLA eligible remained roughly constant between 1998 and 2000, at about one-quarter.

Historically, a small percentage of non-custodial parents have contested the COLA. In 1998, when the state was still operating an administrative process, about 3% of parents (1,357) contested the COLA. Frontline IV-D workers were responsible for completing and filing the paperwork associated with the appeal. In 1999, the Minnesota Supreme Court ruled that the IV-D system had to replace its administrative process with an expedited judicial process to establish and modify orders. As an indirect consequence of that ruling, the responsibility of completing and filing the paperwork transferred to the non-custodial parent. From the non-custodial parent’s perspective, the change in the process made contesting the COLA more time intensive.

Coincident with the change in process, the COLA appeal rate fell to less than one percent,^[20] which represented a 75% decline in appeals. During the last three years, judges denied between 50% and 60% of the appeals (that is, they enacted the COLA in full). The fact that this percentage remained roughly constant during the transition from the administrative to the judicial process implies that the extra paperwork is deterring as many strong cases from appealing as weak ones. This suggests that a number of non-custodial parents who have legitimate grounds to appeal a COLA are failing to do so. State officials also point to the state’s robust economy as a factor in the declining appeal rate.

Minnesota routinely tracks the effect of the COLA on obligation and collection levels. The state does so by comparing obligations/collections for the COLA cases before and after the COLA-effective data (May 1^st ). In 1998, the state estimated that total monthly obligations—among the COLA cases—increased by $678,498, or about $16.45 per case. For the same period, the state estimates monthly collections rose by $844,527, or $20.47 per case. Minnesota has not formally analyzed how the COLA process affects the relationship between awards, state guidelines, and the non-custodial parent’s ability to pay support.

To calculate the impacts reported in Exhibit II.4 , Minnesota analyzes only those cases that received a COLA in a given year. For the selected cases, staff compare obligations and collections before and after the policy’s effective date. Not surprisingly, given the nature of the policy, the staff consistently find that obligations and collections increase after the effective date. However, Minnesota’s method may overstate the impact somewhat because other factors—such as new enforcement tools and a strong economy—may also improve collections. One way to determine whether these other factors play an important role is to analyze some non-COLA cases over the same time period. For example, we may find that—among COLA cases—average monthly collections increased 15% after the effective date. However, an analysis of non-COLA cases may show that their average monthly collections rose 5% during the same period. This latter finding would suggest that factors other than the COLA are simultaneously working to increase collections. We would calculate the COLA’s net effect by subtracting the change measure for the non-COLA cases (that is, 15% minus 5% equals 10%).

As part of this study, Minnesota’s IV-D agency agreed to perform such an analysis using actual data from 1999-2000. Because Minnesota applies the COLA adjustment on a biennial basis, the cohort that is not affected by the COLA in a particular year effectively serves as our non-COLA control group. For example, in May 2000, Minnesota did not adjust orders for the 45,373 cases that had been adjusted in May 1999. So, our analysis compares obligation and collections outcomes in the year 2000 between the 1999 COLA cohort (the control group) and the 2000 COLA cohort (the experimental group). Exhibit II.5 reports the state’s findings.^[21] which represented a 75% decline in appeals. During the last three years, judges denied between 50% and 60% of the appeals (that is, they enacted the COLA in full). The fact that this percentage remained roughly constant during the transition from the administrative to the judicial process implies that the extra paperwork is deterring as many strong cases from appealing as weak ones. This suggests that a number of non-custodial parents who have legitimate grounds to appeal a COLA are failing to do so. State officials also point to the state’s robust economy as a factor in the declining appeal rate.

Minnesota routinely tracks the effect of the COLA on obligation and collection levels. The state does so by comparing obligations/collections for the COLA cases before and after the COLA-effective data (May 1^st ). In 1998, the state estimated that total monthly obligations—among the COLA cases—increased by $678,498, or about $16.45 per case. For the same period, the state estimates monthly collections rose by $844,527, or $20.47 per case. Minnesota has not formally analyzed how the COLA process affects the relationship between awards, state guidelines, and the non-custodial parent’s ability to pay support.

To calculate the impacts reported in Exhibit II.4 , Minnesota analyzes only those cases that received a COLA in a given year. For the selected cases, staff compare obligations and collections before and after the policy’s effective date. Not surprisingly, given the nature of the policy, the staff consistently find that obligations and collections increase after the effective date. However, Minnesota’s method may overstate the impact somewhat because other factors—such as new enforcement tools and a strong economy—may also improve collections. One way to determine whether these other factors play an important role is to analyze some non-COLA cases over the same time period. For example, we may find that—among COLA cases—average monthly collections increased 15% after the effective date. However, an analysis of non-COLA cases may show that their average monthly collections rose 5% during the same period. This latter finding would suggest that factors other than the COLA are simultaneously working to increase collections. We would calculate the COLA’s net effect by subtracting the change measure for the non-COLA cases (that is, 15% minus 5% equals 10%).

As part of this study, Minnesota’s IV-D agency agreed to perform such an analysis using actual data from 1999-2000. Because Minnesota applies the COLA adjustment on a biennial basis, the cohort that is not affected by the COLA in a particular year effectively serves as our non-COLA control group. For example, in May 2000, Minnesota did not adjust orders for the 45,373 cases that had been adjusted in May 1999. So, our analysis compares obligation and collections outcomes in the year 2000 between the 1999 COLA cohort (the control group) and the 2000 COLA cohort (the experimental group). Exhibit II.5 reports the state’s findings.^[22]

Focusing on TANF cases, the analysis indicates that average obligations associated with the 2000 COLA cohort increased by 5.1% between April and June 2000, which is consistent with the actual COLA adjustment in May 2000 (5.3%). Average obligations for the 1999 COLA cohortremained roughly constant.^[23] Over the same period, average collections associated with the 2000 COLA cohort rose 12.0%; however, average monthly collections for the 1999 COLA cohort also increased (6.6%). Because the 1999 COLA cohort was not eligible for the COLA in May 2000, we attribute its 6.6% increase to other child-support policies, enforcement tools, or the economy.^[24] To calculate the policy’s net effect on collections, we subtract the percentage increase found for the 1999 COLA cohort from the effect found for the 2000 COLA cohort—that is, 12.0% minus 6.6%, or 5.4%.

The analysis of non-TANF cases yields similar findings. As with the TANF cases, the 2000 cohort’s obligations and collections increased at a higher rate than that of the 1999 cohort. Notably, however, obligations for the 2000 COLA cohort rose by somewhat less than the May 2000 COLA rate of adjustment (3.2% observed versus 5.3% actual adjustment), which we are unable to explain. Collections for the 2000 cohort increase at about twice the rate as they do for the 1999 COLA cohort, suggesting a net impact of 5.3% (11.0% minus 5.7%).

Caseworkers rarely hear complaints about the COLA from custodial parents. In fact, workers in the Ramsey County office noted that the only custodial parents they hear from are those with out-of-state orders who are not eligible for the COLA and are calling to request it. If custodial parents believe circumstances have changed considerably, they continue to have access to a traditional review and adjustment. As discussed previously, Minnesota modified 3,825 orders in 1998 through the traditional process.

According to the Resource Center for Fathers and Families, the COLA policy erroneously assumes that all parents are doing well because the economy is strong. Some non-custodial parents are barely able to maintain a living, particularly if they have limited education, have recently spent time in prison, or have more than one family to support. The COLA will increase the non-custodial parents’ payments even if their salaries have not kept pace with inflation. Moreover, parents who try to improve their economic situations by going to college are penalized, because increased expenses, such as college loans, are not taken into consideration when determining COLAs. The organization contends that applying COLAs to non-custodial parents whose incomes have not increased may ultimately result in more parents going “underground” and working off the books to avoid payments altogether.

The organization does not oppose the COLA in principle and believes it is important for parents to contribute; some simply cannot do so. Advocates suggest that there should be a way to address the non-custodial parent’s inability to pay support before the COLA is applied. For example, the child support officer could look up the non-custodial parent’s income through the PRISM system to determine whether it justifies a COLA. Caseworkers could also check wages through employment department records (see the Vermont chapter for a more detailed discussion of this issue). Absent these pre-COLA income verifications, the advocates view the procedure as somewhat unfair because once the COLA is processed the non-custodial parent has to spend considerable time determining how to file a motion to stay and then appearing in court.^[25]

Frontline caseworkers are proponents of the COLA policy. The COLA directly reduces the number of formal modifications, which in their view is cumbersome and cost-ineffective. They believe the state runs the COLA process as efficiently as possible and had few recommendations for improvements. Caseworkers did call for a simplified COLA notice, which they believed would reduce the number of inquires they field from non-custodial parents.

According to a Ramsey County (St. Paul) magistrate, the COLA is well supported in the judicial community because the hearings are straightforward (limited in scope to the non-custodial parent’s income) and short. This stands in stark contrast to the standard hearing for reviews and adjustments. The court administrator for Washington County (Stillwater) concurs with this view. While COLA appeals increase activity in April, May and June, she believes that the COLA reduces the court’s overall workload.