A r c h i v e d  I n f o r m a t i o n

Background

Chapter 1 provides financial assistance to school districts to expand and improve instructional programs to meet the needs of educationally disadvantaged children. The study analyzes an array of options for modifying the current provisions for allocating Chapter 1 funds in order to better meet the program's goals, including options for increasing the targeting of Chapter 1 funds to high-poverty districts and schools, improving the timeliness of poverty data used to allocate funds, adjusting for differences in the cost of education, adjusting for differences in fiscal capacity and fiscal effort, and modifying Chapter 1 provisions concerning the comparability of state and local resources. Where available data permit, the report presents analyses of simulations of FY 1993 allocations to states and counties under each formula alternative.

This study was prepared in support of the National Assessment of Chapter 1. It elaborates upon the findings reported in Reinventing Chapter 1: The Current Chapter 1 Program and New Directions.

Key Findings

Targeting Chapter 1 Funds on the Highest-Poverty Communities

Chapter 1 funds are currently spread thinly across almost every county, 93 percent of all school districts, and 71 percent of all public elementary schools. Concentration Grants, intended to supplement Basic Grants in districts with high concentrations of poverty, go to 66 percent of all districts. Almost half of the elementary schools serving fewer than 10 percent poor children participate in Chapter 1, yet 14 percent of high-poverty schools go unserved. If Chapter 1 is to break the link between school poverty and low student achievement, more funds need to be targeted on the schools with the greatest need.

Several strategies are available to achieve roughly the same increase in targeting on high-poverty communities. For example, a goal of targeting half of all Chapter 1 funds to the quartile of counties with the highest poverty rates (which contain 45 percent of the nation's poor children) could be achieved in four different ways:

• Increasing the share of funds allocated through the Concentration Grant provision from 11 percent to 75 percent. This option would strengthen targeting by working within the existing law. However, major increases in the Concentration Grant share are necessary to achieve a substantial increase in targeting to the poorest counties.

• Targeting Concentration Grants more intensively on high-poverty counties while raising the share of funds allocated through Concentration Grants. Raising the eligibility threshold from 15 percent poverty to 18 percent (the national poverty rate for school-age children) while also raising the Concentration Grant share to 50 percent would achieve the goal of targeting half of the funds to the highest poverty quartile of counties.

• Replace Basic and Concentration Grants with an "absorption" formula that assumes that communities can absorb the costs of serving the special needs of children below a poverty threshold of 6 percent. Poor counties would have a greater portion of children remaining in the formula. This option would eliminate 103 low-poverty counties from Chapter 1.

• Replace Basic and Concentration Grants with an "absorption" formula that fully counts students over an 18 percent poverty threshold and weights students below the threshold by one-half. No counties would be eliminated from Chapter 1. This approach would require communities with low poverty rates to shoulder a greater share of the cost for these students, thus shifting part of available funds to more needy communities.

Improving the Timeliness of Poverty Data Used to Allocate Funds

Because Chapter 1 funds are allocated to counties based on decennial census data, areas that experience large demographic shifts over the course of the decade may be proportionally overfunded or underfunded until new census data are released. Options for improving the timeliness of child poverty data include working with the Census Bureau to obtain biennial poverty updates, aging the decennial census data using percentage change in state enrollment, using counts of students eligible for subsidized school lunches, and allowing states to update poverty counts using the best data available.

• Working with the Census Bureau to obtain biennial updates of child poverty counts. The Census Bureau has proposed to develop methods for updating county-level income and poverty estimates every two years. These estimates would combine data from the decennial census, annual federal income tax returns, and annual Census Bureau surveys. The proposed project is experimental, and the reliability of the updated poverty estimates will remain uncertain until they are produced and tested. Although the Census Bureau cannot guarantee reliable county-level child poverty estimates until further research is funded, the proposal provides a promising alternative to decennial poverty counts for a relatively low cost (projected at $420,000 annually). Furthermore, if county-level estimates prove infeasible, the project would at least be able to provide state-level updates that capture inter-state shifts in the distribution of poor children.

• Aging the decennial census data using percentage change in state enrollment. Using this adjustment assumes that the percentage of poor children in each state remains constant, whereas use of the unadjusted census counts assumes that the absolute number of poor children in each state remains constant. If this option had existed during the 1980's, funds would have shifted from states with slow-growing or declining populations to states whose populations grew at above-average rates. A comparison with the 1990 census state-level counts of poor children reveals that the 1980 counts adjusted for percentage change in state enrollment are a more accurate predictor of 1990 poverty rates than is continued use of the unadjusted 1980 census.

• Using counts of students eligible for subsidized school lunches. Because the subsidized lunch program uses a looser definition of poverty — up to 130 percent of the poverty line for free lunch and up to 185 percent for reduced-price lunch — using this measure could reduce the targeting of Chapter 1 funds on the poorest children. In addition, subsidized lunch counts exclude many poor children, particularly high school students and children of illegal aliens. Free lunch participants are distributed very differently among the states than are poor children counted in the census, with the state ratio of 1989 free lunch participants to 1990 census child poverty counts ranging from 184 percent in Hawaii to 106 percent in Montana. Shifting to this poverty measure could cause funding reductions of up to 26 percent in some states.

• Allowing states to update poverty counts using the best data available. Allowing states to disregard the county allocations and distribute funds to LEAs solely on the basis of the poverty measure that they deem most current and reliable would allow adjustments for within-state poverty shifts over the course of the decade. However, it would not address shifts in the distribution of poor children across states.

Adjusting for Differences in the Cost of Education

The current Chapter 1 formula attempts to compensate for differences in the cost of education across states by adjusting county allocations using state average per pupil expenditure (limited to within 80-120 percent of the national average). In FY 1993 the per-pupil expenditure factor redistributed $368 million, or 6 percent of Chapter 1 funds, mainly away from relatively low-income states in the South to relatively high-income states in the Northeast. The state per-pupil expenditure factor has been criticized as an inaccurate cost proxy, because it tends to underestimate costs in low-income, low-expenditure states, so that those who are needier to begin with get less federal help.

Alternatives examined include eliminating the cost factor, restricting the bounds to 90-110 percent of the national average, and substituting an alternative cost index based on average teacher salaries or private-sector wages.

• Eliminating the cost factor. This option would shift funds mainly away from high-income northeastern states to low-income southern states, but would fail to recognize the legitimacy of documented cost differences among states. One recent study estimated that there was a 57 percent variation in purchasing power between the highest- and lowest-cost states in 1990 (McMahon and Chang, 1991). Moreover, because the national poverty line does not adjust for regional differences in the cost of living, the current formula already undercounts poor children in states where the cost of living is high. Eliminating the cost factor would cause funding reductions of up to 21 percent in many needy urban counties, including New York City, Trenton, and Boston.

• Limiting the per pupil expenditure factor to 90-110 percent of the national average. This option would generally shift allocations in the same direction as eliminating the cost factor, but shifts would be less severe. Funding losses of 10 percent would occur in 8 states.

• Using index of state average teacher salaries. Based on NEA data, this option would shift $243 million among states, with $198 million, or 81 percent, going to California. Teacher salaries in California reflect the state's high cost of living, but California loses funds under the current cost proxy because its education expenditures are well below the national average.

• Using index of general wages. This alternative would shift $270 million among states, with California again being a major winner, receiving $136 million. Other states receiving increases are a mixture of western and southern states.

Adjusting for Differences in Fiscal Capacity and Fiscal Effort

While neither fiscal capacity nor fiscal effort is part of the current formula, other federal and state education funding formulas take these factors into consideration. A fiscal capacity adjustment would distribute Chapter 1 funds inversely to a state's capacity to raise revenues. A fiscal effort adjustment would reward states that spend above average on education relative to their fiscal capacity.

• Fiscal capacity adjustment. Incorporating an index of state average income per pupil relative to the national average would increase the share of funding for the poorest quartile of counties from 41 percent to 44 percent, shifting funds away from high-income states, mainly in the Northeast, to relatively low-income states in the South. This alternative tends to have the opposite effects as the per pupil expenditure (cost) factor.

• Fiscal effort adjustment. An adjustment based on the ratio of state-local revenues for elementary-secondary to state fiscal capacity was found to produce funding gains or losses of more than 20 percent in 14 states; however, this adjustment could be scaled back to produce more moderate shifts. Utah was the largest gainer and Massachusetts the biggest loser, but no clear geographic pattern to the funding shifts was found.

Comparability of Resources

Where state school funding systems are inequitable, Chapter 1 may simply buy services in poor districts that wealthy districts routinely provide to all students through regular funds. Furthermore, current Chapter 1 comparability standards that measure equity in terms of dollars and teacher-student ratios may be missing the factors that are critical to student success. The evidence suggests both subtle and obvious differences in the comparability of resources across high- and low-poverty schools and districts. The report concludes with possible options for promoting greater equity in the allocation of all education funds.

Method of Analysis

The report presents simulation results for states, county poverty quartiles (with counties ranked by poverty and grouped into quartiles containing one-fourth of the nation's children in 1990), and the counties containing the ten largest school districts. In order to show the full impact of each formula alternative, allocations were calculated without the current 85 percent hold-harmless provision, which would phase in major redistributive effects over a number of years. Puerto Rico's allocation was held constant from FY 1992. All other current formula provisions were retained.

Further Information

The full report is available from the Planning and Evaluation Service, Office of Undersecretary, 600 Independence Avenue S.W., Room 4165, Washington, D.C. 20202 or calling (202) 401-0590.

-###-

Return to Elementary and Secondary Education Page

mail to esed@ed.gov

Last update September 1996 (swz).