Analysis of the 2006-07 Budget Bill Legislative Analyst's Office February 2006

Economic Impact Aid

We find that the Economic Impact Aid (EIA) funding formula is outdated, results in district allocations that appear arbitrary and unpredictable, and has recently become unworkable. We recommend the Legislature revise the EIA formula so that (1) district allocations are predictable and meet local needs for serving both poor and English learner students and (2) calculations are based on reliable data. To the extent the Legislature wishes to fund Proposition 98 at the Governor’s proposed level, we recommend redirecting some funding in the budget to help ease districts’ transition to a revised formula.

The 2006-07 Governor’s Budget provides $648 million for the Economic Impact Aid (EIA) program. This funding level represents a $61.3 million, or 10.4 percent increase above the current year.

The EIA program funds school districts to provide compensatory education services to low-performing and English learner (EL) pupils. School districts use this funding for a variety of purposes, including: (1) extra assistance to low-achieving pupils, (2) supplemental instruction services to EL students, (3) training to teachers who instruct EL students, and (4) supplementary materials.

In our Analysis of the 2004-05 Budget Bill, we identified several problems with the formula used to distribute EIA funding. To address these issues, the Legislature passed legislation in 2004-05 (SB 1645, Escutia), and adopted budget bill language in 2005-06 to conduct a study of the EIA program and investigate options for a new formula. The Governor vetoed both proposals. In each veto message he directed a working group-consisting of the Department of Finance, the Office of the Secretary for Education, the California Department of Education (CDE), and our office-to develop options for restructuring the EIA formula. While some initial meetings have taken place, neither the administration nor CDE has actively engaged in the working group process, and no progress has been made towards reforming the formula.

The issues we raised about the EIA formula two years ago are still concerns, and a new problem in obtaining one of the primary data inputs used for the EIA calculation has made the need for reform even more urgent. Below, we discuss five problems with the current EIA formula, and present issues and options for the Legislature to consider in improving the formula.

The EIA Formula Is Outdated and Problematic

Districts’ EIA allocations are calculated using a complex formula based on the number of EL and economically disadvantaged students enrolled in each district. The primary data inputs are district enrollment of (1) EL students, as measured by an annual language census, and (2) students from families below the poverty level receiving California Work Opportunity and Responsibility to Kids (CalWORKs) grants. Poverty data from the 1990 Census also factors into the formula. Under the current formula, poverty is by far the most important factor in determining a district’s allocation.

Our analysis indicates the following problems with the current EIA funding formula.

Formula Does Not Reflect Current School Demographics. The EIA funding mechanism has been in existence for more than 25 years. During this time, the state’s demographics and the needs of the student population have changed dramatically. Figure 1 displays the trends for ELs and CalWORKs students over the past 25 years. As the figure shows, in 1980-81 there were almost twice as many children in CalWORKs as ELs. In contrast, by 2004-05 there were two and one-half times as many ELs as CalWORKs children.

The EIA formula, however, has not been updated to reflect these changing dynamics. Despite the increase in EL students and districts’ need for EL funding, poverty is still by far the most important factor in determining a district’s EIA funding allocation. As a result, districts with large numbers of poor students receive far more funding than districts with large numbers of EL students.

Heavy Emphasis on Poverty Skews Per-Pupil Payments. A close look at EIA allocations for two districts of similar size illustrates the formula’s heavy emphasis on poverty. Figure 2 displays the EIA allocations and the number of EL and CalWORKs students in Oakland and Santa Ana Unified School Districts. In 2004-05, Oakland received $389 for each EL and CalWORKs student in the district based on about 29,000 in the two groups. Santa Ana received $236 for each based on about 45,000 in the two groups. As a result, Oakland received about $660,000 more than Santa Ana, despite having 16,000 fewer targeted students. This sort of discrepancy raises fundamental questions about the adequacy of the current EIA formula.

District Allocations Appear Arbitrary and Unpredictable. The current EIA formula is extremely complex. This complexity results in district allocations that are hard to understand based on underlying district demographics. Two districts with almost identical numbers of both EL and CalWORKs students can receive very different amounts of EIA funds. For instance, Sausalito Elementary School District receives $778 for each of the 83 CalWORKs and EL students in the district (66 and 17, respectively). Chawanakee Joint Unified receives only $373 for each of the 81 students in the two groups (67 and 14, respectively). Given that the number of students needing extra assistance in these two districts are almost identical, the widely differing amounts seem hard to justify.

Additionally, the EIA formula creates unpredictable results from year-to-year, leading to counterintuitive scenarios where districts can receive funding increases when their number of EL and CalWORKs students declines, and decreases when they experience an increase in the number of students in the two groups. This lack of predictability can complicate districts’ annual planning efforts.

CalWORKs Counts May No Longer Be a Good Measure of Poverty. Between 1996 and 2002, the counts of CalWORKs students used in the EIA calculation declined by 45 percent. In contrast, the number of California households whose income was below the federal poverty level declined by 25 percent. The sharp decline in CalWORKs participation rates may not mirror the actual decline in the number of children living in poverty, but rather may reflect other factors, such as changes in CalWORKs program requirements. Thus, using CalWORKs for the EIA formula may not reflect actual district needs for serving poor students. This conclusion is supported by the fact that other measures show significantly higher rates of poverty among K-12 children (as discussed in further detail below).

Change in Data Availability Makes Current Formula Unworkable. Until recently, CDE relied on the Department of Social Services (DSS) to provide it with child-specific counts of children in families below the poverty level receiving CalWORKs assistance. The CDE then used this information to derive district-level CalWORKs student counts, one of the primary inputs for the EIA formula. Beginning in December 2004, however, DSS stopped providing CDE with child-specific CalWORKs data. According to DSS, these changes were in response to concerns about the security of confidential data. Without access to this data, CDE is unable to use the current formula to calculate district EIA funding. The CDE handled this issue in the current year by reusing prior-year CalWORKs data for a second consecutive year. This is not a viable long-term solution. It would be inappropriate to use this outdated data for a third time in 2006-07. We believe this problem should be addressed in the budget year.

Legislature Has Several Options for Addressing CalWORKs Data Issues

There are three main paths the Legislature could take to address the CalWORKs data availability problem. It could: (1) try to resolve data issues with DSS to make CalWORKs data available again, (2) eliminate the need for CalWORKs data by using a different measure of poverty in the EIA formula, or (3) remove the poverty measure from the formula altogether and make EIA a program for EL students. Below, we discuss some advantages and disadvantages of each approach.

Option 1: Try to Resolve Data Issues With DSS. The Legislature could investigate the feasibility of requiring DSS to revise its procedures so that CDE has access to the CalWORKs data necessary to compute EIA funding levels. The CDE has been working with DSS for the past year trying to resolve this issue and has determined it is not possible to obtain the necessary data using alternate methods. It is possible, but not certain, that statutory language directing DSS to collect and provide the data could address some of the problems. However, even if DSS can ultimately provide the data, this solution would not address the remaining concern that CalWORKs counts may not accurately reflect the actual number of poor children in a district.

Option 2: Use a Different Measure of Poverty. To address both the availability and accuracy concerns related to CalWORKs data, the Legislature could choose to use a different indicator of poverty for the EIA calculation. We believe there are three alternate data sources the Legislature could consider. Under a new formula, these data could also be used in combination.

• Title I, Part A: Using biannually updated Census data, the federal government allocates funding for this supplemental program based on counts of poor, neglected, delinquent, and foster children. Since this data is used to distribute $1.8 billion in federal Title I funds, districts are generally familiar with the data, and have some comfort level with using it as a measure of poverty. Since the data is collected at the federal level, it can not be manipulated by local decision-making. There have been some concerns raised that the Census may undercount poor children in remote rural areas of the state, but as long as the state maintains a minimum EIA grant size, this may not be a significant problem.

• Free or Reduced Price Meals (FRPM): Schools report the number of students who qualify (based on family income) and enroll to receive federally subsidized meals at school. The FRPM counts do not include those students who qualify but elect not to participate in the program. Moreover, participation in FRPM is skewed toward elementary grades, so economically disadvantaged high school students are likely under-represented in these counts.

• Parental Education Level: For purposes of the federal school accountability program, CDE combines the number of pupils for whom neither parent has completed high school, in conjunction with FRPM participation data, to determine whether the student is socioeconomically disadvantaged. (A low education level is considered an indicator for low-income status.) These data can be problematic in that they are self-reported (in some cases by students, not parents), and are only collected for children in grades 2 through 11.

There are advantages and disadvantages to each of these alternate data sources. Furthermore, because they define “poverty” in different ways, they can result in dramatically different counts of “poor students” within the same district. Figure 3 illustrates examples of these differences, displaying CalWORKs, FRPM, and Title I participation counts within a sample of districts. (Parental education data was not available for this analysis.)

The figure shows that in Coachella Valley Joint Unified, a rural school district, the participation rate for CalWORKs is almost ten times less than for FRPM and four times less than for Title I. This significant discrepancy is also evidenced in Capistrano Unified, a suburban district with relatively low levels of poverty. Statewide totals for each program indicate FRPM and Title I participation rates are significantly higher than CalWORKs. Additionally, the figure shows there can be different trends among the data even between similar districts. Oakland and Los Angeles Unified School Districts are both large urban districts with relatively the same number of poor children as defined by Title I; however, Oakland has a higher participation rate in CalWORKs and Los Angeles has a higher participation rate in FRPM.

These examples reiterate that CalWORKs may not be reflective of the actual number of students living in poverty. The data from the sample districts also highlight that because of the considerable differences among student counts, switching from CalWORKs to an alternate measure of poverty would result in a redistribution of EIA funding across the state.

Option 3: Remove Poverty Measure From EIA Formula. A different solution for the CalWORKs data problems would be to eliminate the poverty factor from the EIA formula altogether, and base allocations solely on counts of EL students. Districts report they use around 85 percent of EIA funds on EL services, so the Legislature could opt to transform EIA into an EL program. To understand why the districts direct most of the EIA funds for EL students, it is necessary to look at resources available from other sources. The federal government provides around $1.8 billion to the state for base Title I grants targeting poor students, and around $275 million for Title III targeting EL students. Given the significantly higher funding for economically disadvantaged students, it makes sense that many districts direct their EIA funds mainly toward serving EL students. Considering EIA in combination with funds from other sources, it may make sense to distribute EIA funds based on EL counts. However, relying solely on EL counts would substantially change the EIA funding distribution and may not recognize the multiplicative difficulty of serving EL students who are also poor.

Simplify Formula and Provide Funding to Ease Transition

The aforementioned problems obtaining CalWORKs data mean that action will likely be needed to address the EIA funding formula in this year’s budget discussions. We recommend the Legislature use this opportunity to undertake a comprehensive reform of the formula, to address both the data availability issue and the numerous other problems we have discussed. In undertaking this revision, we think there are three major issues that are essential to address.

We believe the new EIA formula should be (1) predictable from year-to-year so districts can plan ahead; (2) transparent, so districts and the public understand the rationale behind the funding levels; and (3) calculated using a dependable data source so the formula will remain consistent from year to year.

In addition to the option of replacing or removing CalWORKs as the poverty indicator, the Legislature could also consider various other reforms. We summarize these options for revising the EIA formula in Figure 4 and describe them in further detail below.

Align EIA With Other Programs. There are several other programs that provide supplemental funding to help schools serve economically disadvantaged students (including the federal Title I, Part A program and the state Targeted Instructional Improvement Block Grant) and EL students (including the federal Title III and the state English Learner Acquisition programs). The Legislature may want the revised EIA formula to in some way account for the funding districts receive from these other programs, reconstituting EIA as a component of a larger, coordinated funding package designed to help serve needy students.

Reconsider Formula’s Data Inputs and Weights. As discussed, the Legislature has several options for replacing CalWORKs as the measure of poverty in the EIA formula, including eliminating the poverty factor from the calculation altogether. We recommend the Legislature also reconsider the “weight,” or importance of the two primary EIA factors-EL and poverty status. The Legislature may want to consider changing the weights in the new formula so that poverty is not so heavily weighted, especially given that the EL population has increased significantly since the program was created and districts report using the majority of EIA funds for EL services. Adjusting these funding weights could also be part of the process of aligning EIA with other programs, as discussed above.

Distribute Funds Through Combination of Grant Types. The current EIA formula is structured around three elements: (1) per-pupil grants, under which every district receives a set amount per targeted student, (2) concentration grants, which provide additional funding to districts with a large proportion of poor and EL students, and (3) minimum grants, which provide a minimum level of funding to districts with small numbers of eligible pupils so they have sufficient resources to operate a program. We recommend the Legislature consider maintaining this multi-grant structure, which ensures that districts receive a level of funding that is proportional to the needs of the targeted population.

Provide Transitional Funding. Revising the EIA formula will likely affect the statewide distribution of funds across districts. As such, we recommend the Legislature provide some ongoing funds to hold districts’ funding levels harmless and help ease the transition to the new formula. For instance, the Governor’s budget is already proposing to provide a 10.4 percent year-to-year increase in statewide EIA funding. This would provide some additional funds beyond statutory growth and cost-of-living adjustments that could be directed to help ease the transition for districts whose funding levels would decline under the revised formula. These funds could also be redirected from savings we have identified elsewhere in Proposition 98 spending (see, for example, the “Instructional Materials” section of this chapter).

Conclusion

We believe the Legislature needs to take some action to address the EIA/CalWORKs data issues in the budget year. The time is right, however, for comprehensive reform of the EIA formula. Our office will provide the Legislature with several options during budget hearings, using the guidelines outlined above.