GAO-02-242, Title I Funding: Poor Children Benefit Though Funding Per Poor Child Differs


This is the accessible text file for GAO report number GAO-02-242 
entitled 'Title I Funding: Poor Children Benefit Though Funding Per 
Poor Child Differs' which was released on January 31, 2002. 

This text file was formatted by the U.S. General Accounting Office 
(GAO) to be accessible to users with visual impairments, as part of a 
longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States General Accounting Office: 
GAO: 

Report to Congressional Addressees: 

January 2002: 

Title I Funding: 

Poor Children Benefit Though Funding Per Poor Child Differs: 

GAO-02-242: 

Contents: 

Letter: 

Results in Brief: 

Background: 

Allocations Based on Poverty, but Complex Allocation Process Results in 
Different Allocations Per Poor Child Among States, Districts, and 
Schools: 

Allocations Do Not Fully Adjust to Geographic Shifts in the Number of 
Children From Low-Income Families: 

Allocations Do Not Encourage States To Target Their Own Funds To 
Children From Low-Income Families: 

Using Less Restrictive Hold-Harmless Provisions, Revising Grant 
Formulas, or Raising Eligibility Threshold Could Shift Funds Toward 
Poorer Districts and/or Reduce Funding Variations Among Districts: 

Concluding Observations: 

Agency Comments: 

Appendix I: Objectives, Scope, and Methodology: 

Appendix II: Supporting Data: 

Variation in Funding Per Child Under Alternative Hold-Harmless Rules by 
Poverty Groups: 

Replacing State Per-Pupil Expenditure Factor With a Cost of Education 
Factor: 

Appendix III: Selected Provisions of the “No Child Left Behind Act of 
2001” and Related Appropriations Appendix IV Comments From the 
Department of Education: 

Appendix V: Comments From the U.S. Department of Agriculture: 

Appendix VI: GAO Contacts and Staff Acknowledgments: 

GAO Contacts: 

Staff Acknowledgments: 

Tables: 

Table 1: Distribution of Title I Basic Dollars to States, 1999-2000 
School Year: 

Table 2: Distribution of Basic and Concentration Grant Allocations 
Among Urban and Rural School Districts, 1999-2000 School Year: 

Table 3: Distribution of Title I Schools, Title I Funds, and Children 
in Title I Schools by School’s Percentage of Poor Children, 1999-2000 
School Year: 

Table 4: Use of Poverty Updates in Title I Formula Calculations: 

Table 5: Percentage Differences in Average Funding Variation Under 
Alternative Hold-Harmless Rules Compared With a 100-Percent Rule: 

Table 6: Funding Per Poor Child Under Current Law and Under a Policy 
Option That Funds Targeted Grants Instead of Concentration Grants: 

Table 7: Funding Per Poor Child Under Current Law and Formula Options 
That Replace State Per-Pupil Expenditure Factor With a Teacher Cost 
Factor: 

Table 8: Number of Eligible School Districts, Children, and Funding Per 
Child Under Alternative Eligibility Thresholds: 

Table 9: Number of Ineligible Districts, Poor Children, and Title I 
Funding They Would Receive Under Alternative Eligibility Thresholds: 

Table 10: Distribution of Districts, Poor Children, and Title I Dollars 
by Extent of Difference Between District Allocations and Formula 
Calculations: 

Table 11: Percentage of Poor Children and Percentage of Dollars 
Allocated to Districts Whose Allocations Differed From Formula 
Calculations by More Than 1.5 Percent: 

Table 12: Percentage of Poor Children and Percentage of Dollars 
Allocated to Districts Whose Allocations Differed From Formula 
Calculations by More Than 10 Percent: 

Table 13: Differences Between Actual Allocations and Formula Calculated 
Amounts, 1999-2000 School Year: 

Table 14: Number of Students, Districts, Districts Sampled, and 
Response Rates by Survey Strata: 

Table 15: Sampling Errors: 

Table 16: District Funding Per Child by Poverty Group: 

Table 17: Variation in Funding Per Child Within Poverty Groups: 

Table 18: Percentage Change in Funding With No Hold-Harmless, Analysis 
by Changes in State Poverty 2001-2002 (States Ranked by Percentage 
Growth in the Number of Poor Children): 

Table 19: Funding Per Child by Poverty Group Under Formula Options That 
Replace State Per-Pupil Spending With a Cost of Education Factor: 

Table 20: Variation in Funding Per Child Within Poverty Groups Under 
Formula Options That Replace State Per-Pupil Spending With a Cost of 
Education Factor: 

Table 21: Change in Title I Appropriations Between Fiscal Years 2001 
and 2002: 

Figures: 

Figure 1: Distribution of School Districts by Size and Percentage of 
Children in Poverty: 

Figure 2: Basic Grant Dollars Per Poor Child in Districts With 100 to 
250 Children Eligible for Free or Reduced-Price Lunch, 1999-2000: 

Figure 3: Concentration Grant Dollars Per Poor Child in Districts with 
100-250 Children Eligible for Free or Reduced-Price Lunch, 1999-2000: 

Figure 4: Title I Allocation Priorities of School Districts by 
Location: 

Figure 5: Variation in Funding Per Child Under Alternative Hold-
Harmless Rules (Districts Grouped by Percentage of Poor Children): 

Figure 6: Percentage Change in Funding With No Hold-Harmless Rule 
(States Ranked by Percentage Growth in the Number of Poor Children): 

Figure 7: Comparison of the Variation in Funding Per Child Across 
School Districts Under Formula Options That Replace State Per-Pupil 
Spending With a Cost of Education Factor (Fiscal Year 2001): 

Abbreviations: 

GAO: General Accounting Office: 

NSLP: National School Lunch Program: 

TANF: Temporary Assistance for Needy Families: 

[End of section] 

United States General Accounting Office: 
Washington, DC 20548: 

January 31, 2002: 

Congressional Addressees: 

Title I is the largest federal program supporting elementary and 
secondary education and with annual expenditures of about $8 billion, 
accounts for about 3 percent of total education expenditures 
nationally.[Footnote 1] While state and local funds account for over 90 
percent of national education expenditures, Title I is an important 
source of funding for many high-poverty districts and schools. Created 
in 1965 as part of the War on Poverty, Title I is designed to help 
educate disadvantaged children—those with low academic achievement 
attending schools serving high-poverty areas. 

The amount of funds schools receive is the result of a multi-step 
process that combines formula calculations and state and district 
decisions. States receive funding for their districts from the U.S. 
Department of Education (Education), which calculates how much states’ 
school districts are entitled to based on their numbers of children 
from low-income families (poor children) as measured by Census Bureau 
data and their state’s per-pupil education expenditures (a proxy 
intended to reflect cost differences among states in providing 
education). When states receive Title I funds, they make adjustments to 
the calculated district-level amounts in order to set aside funds for 
state administration and account for differences between their actual 
school districts, including charter schools, and the districts 
appearing in Education’s database.[Footnote 2] Once districts receive 
funds from their states, they have some flexibility in how they 
allocate funds to individual schools, but generally must target schools 
with higher percentages of poor children. Not all school districts 
receive Title I funds because they might not have a minimum number 
and/or percentage of poor children to meet eligibility thresholds. 

In the past several years, the Congress has been concerned about the 
achievement gap between disadvantaged students and their more 
advantaged peers and how to improve the performance of children from 
low-income families. Some studies have indicated that schools with 
higher numbers and percentages of poor children may have higher costs 
associated with raising student performance. The Congress has had a 
dilemma of wanting to improve the performance of poor children through 
increased targeting of poor children while simultaneously protecting 
districts whose share of these children has declined from losing a 
significant amount of funds. To increase targeting, in 1994 the 
Congress added a new formula to the Title I program, which would have 
provided more funds to districts with higher numbers and percentages of 
children from low-income families, but until fiscal year 2003, no funds 
were appropriated for this formula. To protect districts from a 
significant loss of funds when their numbers of poor children decline, 
the Congress has implemented several “hold-harmless” provisions. The 
issues of targeting and hold-harmless provisions were much debated in 
Congress’ recent efforts to reauthorize the Title I program. 

In light of these issues, the Congress mandated GAO to study Title I 
allocations, specifically (1) the extent to which Title I funds are 
allocated to states, school districts, and schools with the greatest 
numbers and percentages of school-age poor children; (2) the extent to 
which allocations of such funds adjust to shifts in the numbers of poor 
children; (3) the extent to which the allocation of Title I funds 
encourages the targeting of state funds to school-age poor children; 
and (4) what options might improve targeting of funds, especially to 
states and school districts with higher numbers and percentages of poor 
children, to more effectively serve those children.[Footnote 3] 

In doing our work, we performed extensive analysis of data collected 
from a number of federal, state, and local sources. We obtained data on 
Title I formula allocations for school years 1999-2000, 2000-01, and 
2001-02 from Education and obtained the actual allocations received by 
school districts in the 1999-2000 school year from state officials. We 
interviewed every state Title I director, surveyed a nationally 
representative sample of school district administrators, and 
interviewed representatives of relevant federal agencies and national 
organizations. We conducted our work from December 2000 through 
December 2001 in accordance with generally accepted government auditing 
standards. Appendix I provides a more detailed discussion of our 
methodology. 

Results in Brief: 

In the 1999-2000 school year, Title I funds were generally targeted 
based on numbers and percentages of poor children, but the complex 
allocation process resulted in differences in actual funding per poor 
child among states, school districts, and schools. States with similar 
numbers and percentages of poor children did not always receive similar 
Title I allocations. The same was true of school districts. State and 
district funding levels differed because factors other than numbers of 
poor children are included in Education’s formula calculation, for 
example, the amount a state spends on education. At the school level, 
more dollars were targeted to schools with higher percentages of poor 
children. However, funding per poor child still varied at the school 
level, reflecting the flexibility districts have in setting priorities 
and allocating funds to individual schools, such as to target funds to 
elementary schools rather than to middle or high schools. 

When the numbers of children from low-income families shift among 
states, Title I allocations adjust, but not completely. In other words, 
a state whose share of the nation’s poor children changed from one year 
to another would not necessarily see a corresponding change in its 
Title I allocation amount. Principally, two factors account for this 
lack of responsiveness — a lack of current poverty data and various 
hold-harmless provisions. Education uses Census Bureau estimates of 
poor children, which are available only on a lagged basis, to calculate 
allocations. Over the past decade, the Census Bureau has been required 
to increase the frequency with which it updates poverty data, allowing 
the Department of Education to use more current Census data to make 
Title I formula calculations more responsive to shifts in poverty. The 
hold-harmless provisions limit the extent to which Title I funds can 
shift at all, even when shifts in poverty occur. In the absence of 
these provisions, some disparities between Title I formula calculations 
and poverty would remain because the Census data always lags behind the 
years in which it is used for the formula. 

The allocation of Title I funds does not encourage states to target 
their own funds to children from low-income families. Our review of the 
Title I statute and regulations found no monetary, statutory, or 
regulatory incentive for them to do so. For example, states do not 
receive extra Title I funding in return for targeting state funds to 
poor children. According to recent interviews with state Title I 
directors and our previous studies, Title I allocations are rarely used 
by states as a model for targeting their own funds. 

A number of policy choices could increase the extent to which Title I 
funds are allocated to states and school districts with high numbers 
and percentages of poor children, if desired. The policy choices for 
doing so include changing the appropriations hold-harmless provisions, 
funding the targeted grant, using an alternative cost factor, and 
raising the basic grant eligibility threshold. Using less restrictive 
hold-harmless provisions would reduce the differences in funding among 
school districts with similar numbers and similar percentages of poor 
children and allocate more to states with more rapidly growing numbers 
of poor children. Funding targeted grants and raising the eligibility 
threshold would shift funding toward districts with higher percentages 
of children in poverty and away from districts with lower percentages. 
Using an alternative cost factor instead of per-pupil expenditures 
would also have the effect of shifting funds to districts with higher 
percentages of children in poverty and to a lesser extent to districts 
with large numbers of children in poverty. The extent to which formula 
changes are desired would depend on the desired balance between, among 
other things, making formula allocations reflective of numbers and 
percentages of poor children and making formula allocations relatively 
stable. The full effect of each change would depend on whether other 
changes were made at the same time and on the extent to which states 
later altered the resulting formula amounts before distributing the 
funds to their school districts. 

In written comments on our draft report, the Department of Education 
generally agreed with the reported findings. The U.S. Department of 
Agriculture, in commenting on the draft report, highlighted its 
concerns about the quality of school lunch data, which we used as one 
of our measures of poverty. Although we acknowledge that these data 
have limitations, we believe that our use of school lunch data, in 
combination with Census poverty estimates, was appropriate. 

Background: 

Title I grants are intended to help elementary and secondary schools 
establish and maintain programs that will improve the educational 
opportunities of low-income and disadvantaged children. Title I funds 
are intended to provide instruction and instructional support for these 
disadvantaged children so that they can master challenging curricula 
and meet state standards in core academic subjects. The law does not 
stipulate exactly how Title I funds are to be spent. Instead, Title I 
is an example of flexible funding that local and state educational 
agencies may use as they deem best. 

Title I funds are directed toward states and school districts with 
greater numbers and percentages of poor children regardless of the 
level of funding they receive from state and local sources. Although 
the amounts that states and localities spend on education vary due to 
differing resource bases and funding priorities, Title I funds are not 
intended to compensate for this variation. 

Federal Allocation Process: 

The U.S. Department of Education each year determines the distribution 
of Title I funds according to the mandates of the law. The authorizing 
legislation in effect through the 2001-02 school year provided for four 
different kinds of Title I grants: 

Basic Grants are the primary vehicle for Title I funding and are the 
easiest grants for which school districts can qualify. Districts are 
eligible for basic grants if they have at least 10 poor children and 
the number of poor children is more than 2 percent of the district’s 
school-age children.[Footnote 4] Nationally, about 92 percent of school 
districts (containing over 99 percent of poor children) receive basic 
grants, which accounted for about 85 percent of the Title I funds 
distributed in fiscal year 1999. 

Concentration Grants are somewhat more directed to poor districts than 
basic grants because district eligibility criteria for concentration 
grants are stricter than those for basic grants. Districts are eligible 
for concentration grants if they have more than 6,500 poor children or 
the number of poor children is more than 15 percent of the district’s 
school-age children. Nationally, about 60 percent of school districts 
(containing about 85 percent of poor children) receive concentration 
grants, which accounted for about 15 percent of the Title I funds 
distributed in fiscal year 1999.[Footnote 5] 

Targeted Grants were not funded until fiscal year 2003.[Footnote 6] 
Targeted grants would be directed more to high-poverty states and 
districts; as the number and percentage of poor children in the 
district increase, the targeted grant amount would increase both in 
absolute dollars and proportionally to other districts. A district 
would be eligible for targeted grants if it had at least 10 poor 
children and these children accounted for at least 5 percent of its 
school-age children. In the 2001-02 school year, about 86 percent of 
school districts, containing 99 percent of poor children would have 
been eligible to receive targeted grants had they been funded. 

Incentive Grants were not funded until fiscal year 2003.[Footnote 7] 
Incentive grants would not be distributed on the basis of poverty, but 
would provide additional funds to states that demonstrate high state 
spending relative to their tax base and states that have less disparity 
in funding among their districts. Under this formula, states would 
distribute funds to districts in proportion to the remainder of their 
Title I allocations. 

Title I funds are distributed from the federal government to the 
states, based on data that are measured at the school district and 
state levels. Since school year 1999-2000, for each school district 
meeting eligibility requirements based on numbers and/or percentages of 
poor children, Education has based its formula calculations on the 
number of poor children in the district. (Prior to that time, formula 
calculations were based on counties rather than school districts. The 
change occurred, in part, due to concern that poor school districts in 
otherwise wealthy counties were not receiving the Title I funds they 
needed.) The funding formula for basic and concentration grants 
principally involves multiplying the number of poor children in a 
school district area, as measured by Census and other data sources, by 
the state’s average per-pupil expenditure, although these amounts are 
subject to hold-harmless provisions and a “small state minimum” 
provision. These key elements of the funding formula for basic and 
concentration grants are described below: 

* Poverty: the number of poor children in the school district area, as 
estimated by decennial Census data and updated by the Census using 
statistical modeling techniques. (The poverty threshold for the 2000-01 
school year for a family of four was an annual income of $17,050.) 

* Expenditures: forty percent of the state’s average per-pupil 
expenditure, limited to a minimum of 80 percent and a maximum of 120 
percent of the national average expenditure.[Footnote 8] This measure 
is included as a proxy for education costs. 

* Small state minimum: Each state is guaranteed a minimum level of 
funding, which is the smaller of either one-quarter of 1 percent of the 
overall appropriation, or the average of one-quarter of 1 percent of 
the overall appropriation and the state’s number of eligible students 
multiplied by 150 percent of the national average per-pupil payment. 
[Footnote 9] 

* Hold-harmless provisions: guarantee each state and district a minimum 
level of funding based on past allocations. That is, the allotment 
cannot be less than a specified percentage of the preceding year’s 
allotment. Such provisions are intended to moderate the effects of any 
large year-to-year shifts in program funding, numbers of eligible 
children, or state education spending.[Footnote 10] The hold-harmless 
provisions protect states with declining numbers of poor children from 
reductions in their allocations, but in the absence of increased 
overall funding, this leaves fewer funds available for states and 
districts with growing numbers of poor children. Specific hold-harmless 
rules have changed over time. In the authorizing statute, the hold-
harmless level for the basic and targeted grants is set at 85, 90, or 
95 percent of the prior year’s allocation, depending on the percentage 
of children in the district who are eligible under the formula (formula 
children). (The concentration grant does not have a hold-harmless 
provision in the authorizing statute.) However, for the 1998-99, 1999-
2000, and 2000-01 school years, provisions in the appropriations 
legislation set the hold-harmless level at 100 percent of the past 
year’s allocation.[Footnote 11] This 100-percent rule was modified for 
school year 2001-02. Under this new rule, rather than being guaranteed 
100 percent of their prior year’s funding amount, districts were 
guaranteed the larger of either their previous year’s allotment or the 
amount they would have received had the authorizing statute’s hold-
harmless rules been in effect.[Footnote 12] 

The Title I program was reauthorized in January 2002, with some 
significant changes. See appendix III for details. 

State Role in the Allocation Process: 

A state’s allocation is the sum of the district allocations determined 
by Education. These allocations, however, are not the final amounts 
that a district will receive. The state must adjust the allocations 
determined by Education to: 

* reserve funds for state administration (up to 1 percent of the amount 
allocated to the state) and for school improvement activities (no more 
than 0.5 percent of the amount allocated to the state but no less than 
$200,000) and; 

* account for changes in district boundaries, district consolidations, 
and the creation or existence of special districts, such as charter 
schools or regional vocational/technical schools, that are eligible for 
Title I funds but may not be reflected in Education’s allocations. 

In the case of special districts that meet the basic and concentration 
grant eligibility criteria, the state must ensure that those districts 
receive the Title I funds to which they are entitled. This may involve 
reducing the allocations of districts from which these special 
districts draw children. 

In addition, the statute gives states the flexibility to use 
alternative poverty data, which Education must approve, to redistribute 
Education-determined basic and concentration grant allocations and re-
determine eligibility for “small” districts serving areas with fewer 
than 20,000 total residents. This provision arose out of concerns about 
the quality of Census poverty estimates for small districts. Roughly 79 
percent of all school districts nationally have a total resident 
population of less than 20,000. Currently, nine states use alternative 
data to redistribute allocations determined by Education among their 
small districts. Most of these states use free and/or reduced-price 
lunch data either exclusively or in combination with Census poverty 
data. 

Within District Allocation Process: 

Once funds have been allocated to the district level, districts can in 
turn allocate funds to the schools. Districts have considerable 
discretion— more so than states—in how they allocate Title I funds. 
Districts may use Title I funds for district-level activities—including 
professional development, preschool programs, school improvement 
initiatives, program administration, and parental involvement efforts. 
Districts then generally allocate the remaining Title I dollars to the 
schools. In distributing these dollars, districts are subject to 
several key restrictions. For example, a district must serve eligible 
schools or attendance areas in rank order according to their poverty 
percentage.[Footnote 13] A district must serve those areas or schools 
above 75 percent poverty, including any middle or high schools, before 
it serves any with a poverty percentage below 75 percent. Once all of 
the schools and areas with a percentage above 75 percent have been 
served, the district may serve lower-poverty areas and schools either 
by continuing with the districtwide ranking or by ranking its schools 
below 75 percent poverty according to grade-span groupings (i.e., K-6, 
7-9, 10-12). If a district ranks by grade-span, it can compare the 
school’s poverty percentage to either the districtwide poverty average 
or the poverty average for the respective grade-span grouping.[Footnote 
14] 

Districts are not required to allocate the same per-pupil amount to 
each school, but if they choose not to, they must allocate higher per-
pupil amounts to poorer schools than they allocate to schools with 
lower concentrations of poverty. In addition, districts may apply for 
and receive waivers of any of these allocation rules. The Title I 
statute also requires that school districts provide eligible private 
school children with Title I educational services or other benefits 
that are “equitable” to those provided to eligible public school 
children. The school district provides these services directly to the 
private school children rather than giving funds to the private school 
itself.[Footnote 15] Within the rules, districts may allocate funds to 
schools as they like. 

Distribution of School Districts by Size and Percentage of Poor 
Children: 

School districts differ in the size of their enrollment and their 
percentages of poor children, as shown in figure 1. For example, there 
are about 7,000 school districts with no more than 1,000 children 
enrolled and about 300 districts with more than 20,000 children 
enrolled. Among districts with less than 1,000 children, about half of 
those districts have more than 35 percent of their children eligible 
for free or reduced-price lunch, while the other half have fewer 
children eligible for free or reduced-price lunch. 

Figure 1: Distribution of School Districts by Size and Percentage of 
Children in Poverty: 

[Refer to PDF for image] 

This figure is a stacked vertical bar graph depicting the following 
data: 

Small (the 7174 districts with fewer than 1000 children enrolled): 
0-34% Poverty: 44%; 
35-49% Poverty: 27%; 
50-75% Poverty: 20%; 
75-100% Poverty: 9%; 
Total: 100%. 

Medium (the 7121 districts districts with between 1000 and 20,000 
children enrolled): 
0-34% Poverty: 60%; 
35-49% Poverty: 20%; 
50-75% Poverty: 16%; 
75-100% Poverty: 4%; 
Total: 100%. 

Large: (the 315 districts with at least 20,000 children enrolled): 
0-34% Poverty: 38%; 
35-49% Poverty: 24%; 
50-75% Poverty: 29%; 
75-100% Poverty: 9%; 
Total: 100%. 

Source: GAO Analysis. 

[End of figure] 

Allocations Based on Poverty, but Complex Allocation Process Results in 
Different Allocations Per Poor Child Among States, Districts, and 
Schools: 

On the whole, total Title I grants were allocated to states and school 
districts on the basis of their numbers of children from low-income 
families in the 1999-2000 school year, but individual states and school 
districts received different grant amounts for each poor child. 
Generally, a state’s share of poor children was roughly proportional to 
its share of funds; however, even small differences between the two 
resulted in substantial differences in funding per poor child. School 
districts, like states, as a whole received Title I allocations that 
were in accordance with the numbers of poor children they had enrolled, 
but actual funding per poor child varied among individual school 
districts with similar numbers of poor children. The pattern of Title I 
grant distributions in terms of numbers of poor children among urban 
and rural school districts is to some extent dependent upon the measure 
of poverty that is used. When school districts allocated funds to 
individual schools, they used the flexibility of the Title I program to 
distribute funds where they believed the need was greatest. Even with 
this flexibility, most school districts allocated the majority of funds 
to schools in which at least half of the children were classified as 
poor. 

Title I Allocations Per Poor Child Differ Across States: 

Generally, states with higher numbers of poor children received higher 
amounts of Title I basic grant funds, and states with lower numbers of 
poor children received lower amounts of Title I basic grant funds. 
[Footnote 16] However, even small differences between a state’s share 
of poor children and its share of funds resulted in substantial 
differences in funding per poor child. For example, in the 1999-2000 
school year, Texas had about 10 percent of the nation’s Census poor 
children and received about 9 percent of the Title I basic grant 
dollars while Minnesota had about 1 percent of the poor children and 
1.2 percent of the dollars. However, Texas’s share of the basic grants 
was about 14 percent less than its share of the poor children while 
Minnesota’s share of the basic grants was about 16 percent more than 
its share of the children. As a result, Texas’s basic grants amounted 
to $581 per poor child while Minnesota’s basic grants amounted to $793 
per poor child.[Footnote 17] Table 1 displays for each state and the 
District of Columbia its share of Census poor children, share of basic 
grant dollars, the percentage difference between these shares, and the 
resulting basic grant amount per poor child. 

Table 1: Distribution of Title I Basic Dollars to States, 1999-2000 
School Year: 

Percentage difference between share of poor children and share of basic 
grant funds[B]: 

State: Alabama; 
Share of Census poor children: 2.02; 
Share of basic grant funds[A]: 1.72; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -16.0; 
Basic grant dollars per poor child: %571. 

State: Alaska; 
Share of Census poor children: 0.16; 
Share of basic grant funds[A]: 0.25; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 43.9; 
Basic grant dollars per poor child: $1,033. 

State: Arizona; 
Share of Census poor children: 2.02; 
Share of basic grant funds[A]: 1.61; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -22.6; 
Basic grant dollars per poor child: $561. 

State: Arkansas; 
Share of Census poor children: 1.28; 
Share of basic grant funds[A]: 1.07; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -17.9; 
Basic grant dollars per poor child: $534. 

State: California; 
Share of Census poor children: 14.69; 
Share of basic grant funds[A]: 12.73; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 0-14.3; 
Basic grant dollars per poor child: $581. 

State: Colorado; 
Share of Census poor children: 0.94; 
Share of basic grant funds[A]: 0.97; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 3.1; 
Basic grant dollars per poor child: $686. 

State: Connecticut; 
Share of Census poor children: 0.8; 
Share of basic grant funds[A]: 0.94; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 16.1; 
Basic grant dollars per poor child: $792. 

State: Delaware; 
Share of Census poor children: 0.18; 
Share of basic grant funds[A]: 0.26; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 36.4; 
Basic grant dollars per poor child: $972. 

State: District of Columbia; 
Share of Census poor children: 0.28; 
Share of basic grant funds[A]: 0.34; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 19.4; 
Basic grant dollars per poor child: $826. 

State: Florida; 
Share of Census poor children: 5.63; 
Share of basic grant funds[A]: 4.87; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -14.5; 
Basic grant dollars per poor child: $580. 

State: Georgia; 
Share of Census poor children: 3.28; 
Share of basic grant funds[A]: 2.82; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -15.1; 
Basic grant dollars per poor child: $577. 

State: Hawaii; 
Share of Census poor children: 0.28; 
Share of basic grant funds[A]: 0.28; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 0; 
Basic grant dollars per poor child: $662. 

State: Idaho; 
Share of Census poor children: 0.36; 
Share of basic grant funds[A]: 0.31; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -14.9; 
Basic grant dollars per poor child: $582. 

State: Illinois; 
Share of Census poor children: 3.9; 
Share of basic grant funds[A]: 4.46; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 13.4; 
Basic grant dollars per poor child: $767. 

State: Indiana; 
Share of Census poor children: 1.51; 
Share of basic grant funds[A]: 1.61; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 6.4; 
Basic grant dollars per poor child: $717. 

State: Iowa; 
Share of Census poor children: 0.67; 
Share of basic grant funds[A]: 0.74; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 9.9; 
Basic grant dollars per poor child: $746. 

State: Kansas; 
Share of Census poor children: 0.7; 
Share of basic grant funds[A]: 0.75; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 6.9; 
Basic grant dollars per poor child: $722. 

State: Kentucky; 
Share of Census poor children: 1.79; 
Share of basic grant funds[A]: 1.72; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -4.0; 
Basic grant dollars per poor child: $644. 

State: Louisiana; 
Share of Census poor children: 2.76; 
Share of basic grant funds[A]: 2.57; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -7.1; 
Basic grant dollars per poor child: $623. 

State: Maine; 
Share of Census poor children: 0.33; 
Share of basic grant funds[A]: 0.42; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 24.0; 
Basic grant dollars per poor child: $835. 

State: Maryland; 
Share of Census poor children: 1.16; 
Share of basic grant funds[A]: 1.4; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 18.8; 
Basic grant dollars per poor child: $806. 

State: Massachusetts; 
Share of basic grant funds[A]: 2.01; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 27.8; 
Basic grant dollars per poor child: $885. 

State: Michigan; 
Share of Census poor children: 3.54; 
Share of basic grant funds[A]: 4.42; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 22.1; 
Basic grant dollars per poor child: $838. 

State: Minnesota; 
Share of Census poor children: 1.02; 
Share of basic grant funds[A]: 1.2; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 16.2; 
Basic grant dollars per poor child: $793. 

State: Mississippi; 
Share of Census poor children: 1.67; 
Share of basic grant funds[A]: 1.7; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 1.8; 
Basic grant dollars per poor child: $683. 

State: Missouri; 
Share of Census poor children: 1.93; 
Share of basic grant funds[A]: 1.8; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -7.0; 
Basic grant dollars per poor child: $625. 

State: Montana; 
Share of Census poor children: 0.36; 
Share of basic grant funds[A]: 0.35; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -2.8; 
Basic grant dollars per poor child: $662. 

State: Nebraska; 
Share of Census poor children: 0.36; 
Share of basic grant funds[A]: 0.45; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 22.2; 
Basic grant dollars per poor child: $849. 

State: Nevada; 
Share of Census poor children: 0.39; 
Share of basic grant funds[A]: 0.32; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -19.7; 
Basic grant dollars per poor child: $556. 

State: New Hampshire; 
Share of Census poor children: 0.16; 
Share of basic grant funds[A]: 0.25; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 43.9; 
Basic grant dollars per poor child: $556. 

State: New Jersey; 
Share of Census poor children: 1.92; 
Share of basic grant funds[A]: 2.34; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 19.7; 
Basic grant dollars per poor child: $816. 

State: New Mexico; 
Share of Census poor children: 1.15; 
Share of basic grant funds[A]: 0.89; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -25.5; 
Basic grant dollars per poor child: $520. 

State: New York; 
Share of Census poor children: 8.23; 
Share of basic grant funds[A]: 9.8; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 17.4; 
Basic grant dollars per poor child: $798. 

State: North Carolina; 
Share of Census poor children: 2.45; 
Share of basic grant funds[A]: 1.99; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -20.7; 
Basic grant dollars per poor child: $546. 

State: North Dakota; 
Share of Census poor children: 0.19; 
Share of basic grant funds[A]: 0.26; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 31.1; 
Basic grant dollars per poor child: $897. 

State: Ohio; 
Share of Census poor children: 3.5; 
Share of basic grant funds[A]: 4.11; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 16.0; 
Basic grant dollars per poor child: $786. 

State: Oklahoma; 
Share of Census poor children: 1.62; 
Share of basic grant funds[A]: 1.3; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -21.9; 
Basic grant dollars per poor child: $539. 

State: Oregon; 
Share of Census poor children: 0.81; 
Share of basic grant funds[A]: 0.94; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 14.9; 
Basic grant dollars per poor child: $777. 

State: Pennsylvania; 
Share of Census poor children: 33.59; 
Share of basic grant funds[A]: 4.48; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 22.1; 
Basic grant dollars per poor child: $837. 

State: Rhode Island; 
Share of Census poor children: 0.29; 
Share of basic grant funds[A]: 0.34; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 15.9; 
Basic grant dollars per poor child: $799. 

State: South Carolina; 
Share of Census poor children: 1.68; 
Share of basic grant funds[A]: 1.35; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -21.8; 
Basic grant dollars per poor child: $540. 

State: South Dakota; 
Share of Census poor children: 0.29; 
Share of basic grant funds[A]: 0.26; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -10.9; 
Basic grant dollars per poor child: $612. 

State: Tennessee; 
Share of Census poor children: 1.9; 
Share of basic grant funds[A]: 1.74; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -8.8; 
Basic grant dollars per poor child: $613. 

State: Texas; 
Share of Census poor children: 10.41; 
Share of basic grant funds[A]: 9.01; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -14.4; 
Basic grant dollars per poor child: $581. 

State: Utah; 
Share of Census poor children: 0.41; 
Share of basic grant funds[A]: 0.47; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 13.6; 
Basic grant dollars per poor child: $775. 

State: Vermont; 
Share of Census poor children: 0.14; 
Share of basic grant funds[A]: 0.24; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 52.6; 
Basic grant dollars per poor child: $1,150. 

State: Virginia; 
Share of Census poor children: 1.87; 
Share of basic grant funds[A]: 1.58; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: -16.8; 
Basic grant dollars per poor child: $568. 

State: Washington; 
Share of Census poor children: 1.44; 
Share of basic grant funds[A]: 1.49; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 3.4; 
Basic grant dollars per poor child: $695. 

State: West Virginia; 
Share of Census poor children: 0.9; 
Share of basic grant funds[A]: 0.99; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 9.5; 
Basic grant dollars per poor child: $741. 

State: Wisconsin; 
Share of Census poor children: 1.36; 
Share of basic grant funds[A]: 1.81; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 28.4; 
Basic grant dollars per poor child: $891. 

State: Wyoming; 
Share of Census poor children: 0.14; 
Share of basic grant funds[A]: 0.23; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 48.7; 
Basic grant dollars per poor child: $1,131. 

U.S. total or average: 
Share of Census poor children: 100; 
Share of basic grant funds[A]: 100; 
Percentage difference between share of poor children and share of basic 
grant funds[B]: 6.7; 
Basic grant dollars per poor child: $671. 

[A] Basic grant amounts are the sum of those the state reported 
allocating to their districts and do not include amounts states 
reserved for administration or school improvements. 

[B] Percentage differences calculated as follows: 100* [(poverty 
share - dollar share)/((poverty share + dollar share)/2)]. 

Source: GAO Analysis. 

[End of table] 

The shares of basic grant dollars that states received differed from 
their shares of poor children due to the factors in addition to the 
number of poor children that are included in the federal funding 
formula: state per-pupil expenditures, the small state minimum 
provision and the hold-harmless provision. For example, Vermont, 
Wyoming, New Hampshire, and Alaska, which each received much larger 
basic grants per poor child than the rest of the states, are each also 
among the small-population states that benefited from the small state 
minimum provision in the 1999-2000 school year. Another of the formula 
factors — the state per-pupil expenditures factor — also had an 
important impact on the distribution of Title I dollars to states. 
States with higher percentages of poor children would be expected, all 
else being equal, to have lower tax bases and tend to have lower state 
education expenditures. As a result, states with higher percentages of 
poor children tended to receive smaller basic grant amounts per poor 
child because Title I allocations are based, in part, on the amount a 
state spends on education. We found that of the 17 states with more 
than 20 percent of their children living in poverty in 1997, all but 4 
had basic grant dollars per poor child below the national average in 
the 1999-2000 school year. 

Title I Allocations Were Made in Accordance With Numbers of Poor 
Children, But Amounts Per Poor Child Differed Across School Districts: 

An examination of the allocations that states reported making to their 
school districts in the 1999-2000 school year shows that school 
districts, like states, as a whole received Title I allocations that 
were in accordance with the numbers of poor children they had enrolled, 
but that actual funding per poor child varied considerably among 
individual school districts with similar numbers of poor children. 
[Footnote 18] For example, among districts with between 101 and 250 
children eligible for free or reduced-price lunch, the median basic 
grant funding per poor child was $370, but ranged from $0 to $2,573. 
Concentration grants among school districts with similar numbers of 
poor children also varied. For example, among districts receiving 
concentration grants with between 101 and 250 children eligible for 
free or reduced-price lunch, the median concentration grant funding per 
poor child was $63, but ranged from $1 to $3,547. Figures 2 and 3 
illustrate these findings. The variation illustrated for the national 
level in figures 2 and 3 is similar to the type of variation that 
occurs within states as well. In other words, within states, individual 
districts with similar numbers of children received different 
allocation amounts. 

Figure 2: Basic Grant Dollars Per Poor Child in Districts With 100 to 
250 Children Eligible for Free or Reduced-Price Lunch, 1999-2000: 

[Refer to PDF for image] 

This figure is a vertical bar graph depicting basic grant dollars per 
poor child in districts with 100 to 250 children eligible for free or 
reduced-price lunch, 1999-2000. The vertical axis of the graph 
represents number of school districts. The horizontal axis of the graph 
represents dollars per poor child. 
Standard deviation = 229.57; 
Median = $370; 
Total number of school districts: 3061. 

Source: GAO Analysis. 

[End of figure] 

Figure 3: Concentration Grant Dollars Per Poor Child in Districts with 
100-250 Children Eligible for Free or Reduced-Price Lunch, 1999-2000: 

[Refer to PDF for image] 

This figure is a vertical bar graph depicting concentration grant 
dollars per poor child in districts with 100-250 children eligible for 
free or reduced-price lunch, 1999-2000. The vertical axis of the graph 
represents number of school districts. The horizontal axis of the graph 
represents dollars per poor child. 
Standard deviation = 52.37; 
Median = $63; 
Total number of school districts: 1354. 

Source: GAO Analysis. 

[End of figure] 

Not only did districts with similar numbers of poor children receive 
different allocations per poor child, but also in some cases, school 
districts with larger numbers of poor children received less funding 
per poor child than districts with smaller numbers of poor children. 
For example, in California, one school district with 961 poor children 
received a basic grant of $363 per poor child, while a school district 
with 13 poor children received $1,656 per poor child.[Footnote 19] 
Concentration grants for these two districts varied similarly and 
similar examples of variation occurred in most states. Differences in 
Title I funding per poor child among school districts is due, in part, 
to hold-harmless provisions. 

Allocations Among Urban and Rural School Districts Vary Slightly From 
Their Numbers of Census Poor Children: 

The pattern of Title I grant distributions in terms of numbers of poor 
children among urban and rural school districts in the 1999-2000 school 
year is to some extent dependent upon the measure of poverty that is 
used, as shown in table 2.[Footnote 20] When Census data are used as 
the measure of poverty, both urban and rural school districts as a 
whole received shares of both basic and concentration grants roughly 
proportional to their share of poor children, with some small 
differences.[Footnote 21] However, when eligibility for free or reduced-
price lunch is used as the measure of poverty, urban districts received 
shares of both basic and concentration grants that were greater than 
their share of poor children. 

Table 2: Distribution of Basic and Concentration Grant Allocations 
Among Urban and Rural School Districts, 1999-2000 School Year: 

Basic grants: Urban districts; 
Percentage share of basic grants allocated: 48%; 
Percentage share of Census poor children: 48%; 
Percentage share of children eligible for free or reduced-price lunch: 
44%. 

Basic grants: Rural districts; 
Percentage share of basic grants allocated: 16%; 
Percentage share of Census poor children: 15%; 
Percentage share of children eligible for free or reduced-price lunch: 
16%. 

Basic grants: Other districts; 
Percentage share of basic grants allocated: 36%; 
Percentage share of Census poor children: 37%; 
Percentage share of children eligible for free or reduced-price lunch: 
40%. 

Basic grants: Total: 
Percentage share of basic grants allocated: 100%; 
Percentage share of Census poor children: 100%; 
Percentage share of children eligible for free or reduced-price lunch: 
100%. 

Concentration grants: Urban districts; 
Percentage share of basic grants allocated: 58%; 
Percentage share of Census poor children: 56%; 
Percentage share of children eligible for free or reduced-price lunch: 
53%. 

Concentration grants: Rural districts; 
Percentage share of basic grants allocated: 13%; 
Percentage share of Census poor children: 14%; 
Percentage share of children eligible for free or reduced-price lunch: 
14%. 

Concentration grants: Other districts; 
Percentage share of basic grants allocated: 29%; 
Percentage share of Census poor children: 30%; 
Percentage share of children eligible for free or reduced-price lunch: 
33%. 

Concentration grants: Total; 
Percentage share of basic grants allocated: 100%; 
Percentage share of Census poor children: 100%; 
Percentage share of children eligible for free or reduced-price lunch: 
100%. 

Note: Analysis considered only those districts for which both free and 
reduced-price lunch and Census poverty data were available. Four 
states, Florida, Hawaii, Maryland, and West Virginia, with state or 
county-based school districts, were excluded from the analysis because 
of the imprecision of their location codes. 

[A] Concentration grant analysis includes only those districts that are 
eligible for concentration grants and the children in those districts. 

Source: GAO Analysis. 

[End of table] 

Even small differences between shares of grants and shares of poor 
children resulted in per child dollar differences between urban and 
rural districts. For example, when Census data are used as the measure 
of poverty, rural districts overall had a higher median basic grant 
amount per poor child ($705) than did urban districts ($610). This 
resulted, in part, because the share of basic grants that rural 
districts received was 1 percentage point greater than their share of 
poor children while the share of basic grants that urban districts 
received was equivalent to their share of children. Because the pattern 
of Title I grant distributions among urban and rural school districts 
is dependent on the measure of poverty used, it is worth noting that 
currently all states base their allocations on Census data for areas 
with populations greater than 20,000 residents, and the overwhelming 
majority of states (41) base their allocations on Census data rather 
than subsidized lunch for areas with populations smaller than 20,000 
residents. 

Funding Per Poor Child for Charter Schools Is Less Than That for Other 
Districts Comparable in Size: 

Title I allocations to charter schools that are independent school 
districts are lower than per-pupil allocations to other similarly sized 
school districts.[Footnote 22] Charter schools are not included in 
Education’s Title I formula calculations, but are guaranteed funding on 
an equal basis with other school districts. Consequently, those states 
with independent charter schools must reallocate resources from other 
school districts to these schools. In the 1999-2000 school year, 14 
states and the District of Columbia had operating charter schools that 
were considered independent school districts; 533 of these 734 charter 
schools received basic grants.[Footnote 23] In these states combined, 
charter school districts that received Title I grants had almost 6 
percent of the children receiving free or reduced-price lunch and 
received less than 5 percent of the basic grant funding that was 
distributed to districts with fewer than 2,500 students. Average basic 
grant funding per poor child in the charter school districts was $365 
and was $481 in the other school districts. 

School Districts Use the Flexibility of the Title I Program When 
Allocating Funds to Eligible Schools, and the Majority of Title I Funds 
go to High-Poverty Schools: 

When school districts allocate funds to individual schools, they use 
the flexibility of the Title I program to distribute funds where they 
believe the need is greatest and in ways that they believe best 
provides needed activities efficiently and consistently. Unlike school 
district allocations, which are based on numbers of poor children, 
allocations to individual schools are required by law to be based on 
the percentages of poor children within grade span or within the 
district as a whole.[Footnote 24] From our national survey of school 
districts, we estimate that a majority of school districts (63 percent) 
prioritized Title I funding to eligible primary or elementary grades 
before funding other grade spans, while an estimated 23 percent of 
school districts prioritized funds to schools with higher percentages 
of poor children, regardless of grade span. (See fig. 4.) A similar 
pattern was found among rural districts, of which an estimated 67 
percent targeted funds to needy primary schools. In contrast, equal 
percentages of urban school districts targeted funds to primary schools 
(an estimated 42 percent) and to schools with higher percentages of 
poor children (an estimated 42 percent).[Footnote 25] 

Figure 4: Title I Allocation Priorities of School Districts by 
Location: 

[Refer to PDF for image] 

This figure is a stacked vertical bar graph depicting the following 
data: 

All school districts: 
Primary grades: 63%; 
Poverty ranking: 23%; 
Other priorities: 15%. 

Urban school districts: 
Primary grades: 42%; 
Poverty ranking: 42%; 
Other priorities: 17%. 

Rural school districts: 
Primary grades: 67%; 
Poverty ranking: 17%; 
Other priorities: 16%. 

Note: School district officials self-identified their districts as 
urban, rural, suburban, or mixed. As a result, the urban and rural 
designations in figure 4 are not comparable to those in table 2. 

Source: GAO Analysis. 

[End of figure] 

In addition to distributing funds to individual schools, we estimate 
that over half the districts (58 percent) reserve at least some of 
their Title I grant, typically no more than 10 percent, for 
administration or other district activities. An estimated 61 percent of 
districts that reserve funds for such purposes do so because they 
believe it is more efficient or promotes consistency across schools in 
the district. Districts use the reserved funds for such activities as 
professional development for teachers, parental involvement programs, 
preschool and summer school programs, and before and after school 
enrichment activities.[Footnote 26] 

In addition to the flexibility districts are allowed under the Title I 
regulations, a district may apply for a waiver from the regulations if 
a district finds that it cannot use Title I funds to best serve the 
needs of its schools within the existing regulations.[Footnote 27] 
Waivers are used most commonly to allow districts to (1) serve schools 
of the same grade span without regard for their rank order in terms of 
poverty or (2) allow lower-poverty schools to use Title I funds in 
schoolwide programs rather than for specific students, a practice 
otherwise allowed only in schools with greater than 50 percent of their 
children living in poverty. 

Because districts have flexibility in making Title I allocations to 
schools, and not all schools with poor children receive funds from 
their districts, the amount of Title I funds that individual schools 
receive per poor child differs among schools. Only three states, 
California, Mississippi, and Georgia, were able to provide us with 
electronic information on enrollment, free and reduced-price lunch 
eligibility, and Title I dollar amounts allocated to individual 
schools. Analyses of these three states’ data show that in the 1999-
2000 school year, among the schools receiving Title I funds, schools 
with higher percentages of poor children received a larger percentage 
of the funds than schools with lower percentages of poor children, as 
shown in table 3. Although these findings are based on one year of data 
from only three states, they are consistent with Education’s findings 
that in the 1997-98 school year, nationally 66 percent of schools 
receiving Title I funds had more than 50 percent of their children 
eligible for free or reduced-price lunch, and 35 percent of schools 
receiving Title I funds had at least 75 percent of their children 
eligible for free or reduced-price lunch. 

Table 3: Distribution of Title I Schools, Title I Funds, and Children 
in Title I Schools by School’s Percentage of Poor Children, 1999-2000 
School Year: 

School poverty level: California; 
0-34% poverty: 
Percentage of Title I schools: 15%; 
Percentage of Title I funds allocated to schools: 6%; 
Percentage of poor children in Title I schools[A]: 4%. 

35-49% poverty: 
Percentage of Title I schools: 15%; 
Percentage of Title I funds allocated to schools: 9%; 
Percentage of poor children in Title I schools[A]: 9%. 

50-74% poverty: 
Percentage of Title I schools: 31%; 
Percentage of Title I funds allocated to schools: 29%; 
Percentage of poor children in Title I schools[A]: 30%. 

75-100% poverty: 
Percentage of Title I schools: 39%; 
Percentage of Title I funds allocated to schools: 56%; 
Percentage of poor children in Title I schools[A]: 57%. 

Total: 
Percentage of Title I schools: 100%; 
Percentage of Title I funds allocated to schools: 100%; 
Percentage of poor children in Title I schools[A]: 100%. 

School poverty level: Georgia; 

0-34% poverty: 
Percentage of Title I schools: 6%; 
Percentage of Title I funds allocated to schools: 2%; 
Percentage of poor children in Title I schools[A]: 3%. 

35-49% poverty: 
Percentage of Title I schools: 18%; 
Percentage of Title I funds allocated to schools: 13%; 
Percentage of poor children in Title I schools[A]: 14%. 

50-74% poverty: 
Percentage of Title I schools: 44%; 
Percentage of Title I funds allocated to schools: 43%; 
Percentage of poor children in Title I schools[A]: 44%. 

75-100% poverty: 
Percentage of Title I schools: 33%; 
Percentage of Title I funds allocated to schools: 42%; 
Percentage of poor children in Title I schools[A]: 39%. 

Total 100 100 
Percentage of Title I schools: 100%; 
Percentage of Title I funds allocated to schools: 100%; 
Percentage of poor children in Title I schools[A]: 100%. 

School poverty level: Mississippi; 

0-34% poverty: 
Percentage of Title I schools: 0.4%; 
Percentage of Title I funds allocated to schools: 0.3%; 
Percentage of poor children in Title I schools[A]: 0.2%. 

35-49% poverty: 
Percentage of Title I schools: 9%; 
Percentage of Title I funds allocated to schools: 6%; 
Percentage of poor children in Title I schools[A]: 7%. 

50-74% poverty: 
Percentage of Title I schools: 38%; 
Percentage of Title I funds allocated to schools: 31%; 
Percentage of poor children in Title I schools[A]: 33%. 

75-100% poverty: 
Percentage of Title I schools: 52%; 
Percentage of Title I funds allocated to schools: 63%; 
Percentage of poor children in Title I schools[A]: 59%. 

Total: 
Percentage of Title I schools: 100%; 
Percentage of Title I funds allocated to schools: 100%; 
Percentage of poor children in Title I schools[A]: 100%. 

Note: Amounts may not sum to 100 percent due to rounding. 

[A] Poor children are defined as children eligible for free or reduced-
price lunch. 

Source: GAO analysis. 

[End of table] 

Allocations Do Not Fully Adjust to Geographic Shifts in the Number of 
Children From Low-Income Families: 

When the number of poor children from low-income families shifts 
between states, Title I allocations do not fully adjust in response. 
[Footnote 28] More frequent updates of Census poverty data over the 
past decade have provided Education the data it needs to adjust Title I 
formula calculations to shifts in poverty more quickly than it could in 
the past. However, recent appropriations hold-harmless provisions and 
the small state minimum provision in the formula have limited the 
extent to which Title I allocations can shift at all, even when 
Education has data indicating that shifts in poverty have occurred. 
Even if these rules were changed, allowing formula calculations to be 
based more completely on poverty, the remaining lags in Census poverty 
data would continue to prevent Title I formula calculations from fully 
adjusting to shifts in poverty. 

When Poverty Has Shifted Among States, Allocations Have Not Fully 
Adjusted: 

To the extent that the number of poor children among the states has 
shifted, Title I funding has not completely adjusted in response. For 
example, in 1980, California had 9 percent of the nation’s poor 
children and received 10 percent of all Title I dollars. By 1997, 
California had 16 percent of the nation’s poor children, but received 
just 12 percent of all Title I dollars. This type of disparity has 
occurred in a number of states. In 1997, 33 states received Title I 
allocations that differed from their shares of poor children by more 
than 10 percent. 

Census Is Updating Poverty Data More Frequently: 

A lack of updated poverty data contributed to the mismatch between 
poverty and funding over the last two decades. Education relied on 1980 
decennial Census data to make allocations from 1984 to 1993, at which 
point Education began using 1990 decennial Census data to adjust 
allocations. As shown in table 4, since 1990, the lag in poverty data 
has decreased. This decrease occurred, in part, because in 1994, the 
Congress authorized the Census Bureau to update the data more 
frequently. 

Table 4: Use of Poverty Updates in Title I Formula Calculations: 

Year of decennial poverty data or update: 1979; 
School year(s) in which updated data were used in formula calculations: 
1984-85 through 1992-93. 

Year of decennial poverty data or update: 1989; 
School year(s) in which updated data were used in formula calculations: 
1993-94 through 1996-97. 

Year of decennial poverty data or update: 1993; 
School year(s) in which updated data were used in formula calculations: 
1997-98 through 1998-99. 

Year of decennial poverty data or update: 1995; 
School year(s) in which updated data were used in formula calculations: 
1999-2000 through 2000-01. 

Year of decennial poverty data or update: 1997; 
School year(s) in which updated data were used in formula calculations: 
2001-02. 

Source: GAO Analysis. 

[End of table] 

Hold-Harmless and Small State Minimum Provisions Limit Shifts in Title 
I Allocations Despite Updated Data: 

Although the Census updates provided Education with more timely poverty 
estimates, the hold-harmless and small state minimum provisions limited 
the extent to which Title I formula allocations could change in 
response. These provisions limit the extent to which Title I grants are 
reduced for districts and small-population states with declining 
numbers of poor children. As a result, funds are not sufficient to 
provide “full” formula amounts to districts and states with increasing 
numbers of poor children.[Footnote 29] Because of the effects of the 
hold-harmless and small state minimum provisions, between school years 
2000-01 and 2001-02, holding other things constant,[Footnote 30] most 
states with increased numbers of poor children would have received a 
decreased Title I allocation per poor child, while most states with 
decreased numbers of poor children would have received an increased 
Title I allocation per poor child.[Footnote 31] States with increased 
numbers of poor children would have lost an average of $25 per poor 
child under the basic grant and $10 per poor child under the 
concentration grant.[Footnote 32] States with decreasing numbers of 
poor children between these 2 years would have gained $23 per poor 
child under the basic grant and $4 per child under the concentration 
grant.[Footnote 33] 

While the hold-harmless and, to a lesser extent, small state minimum 
provisions create imbalances between poverty and funding, in the 
absence of these provisions, unavoidable lags in poverty data would 
prevent Education from fully adjusting Title I allocations to shifts in 
poverty. 

Allocations Do Not Encourage States To Target Their Own Funds To 
Children From Low-Income Families: 

Title I allocations do not encourage states to target their own funds 
to children from low-income families. Our review of the Title I statute 
and regulations found no formal monetary, statutory, or regulatory 
incentives for states to target their funds in this way. In our 
interviews with Title I directors in each of the 50 states, only four 
reported that they have programs for disadvantaged children that model 
their eligibility criteria on the Title I program. Furthermore, our 
study[Footnote 34] based on 1991-92 school year data found large 
differences in the extent to which state funding was targeted to school 
districts on the basis of poverty criteria, indicating that states were 
not systematically following the allocation model of the Title I 
program. 

The incentive grant, if funded, could provide an incentive for states 
to spend more of their own dollars. However, the grant would not 
encourage states to target their own funds to children from low-income 
families. In addition, the amount of money that could be provided 
through an incentive grant is not likely to be sufficient to create 
changes in states’ behaviors. 

Using Less Restrictive Hold-Harmless Provisions, Revising Grant 
Formulas, or Raising Eligibility Threshold Could Shift Funds Toward 
Poorer Districts and/or Reduce Funding Variations Among Districts: 

A number of policy options could affect the extent to which Title I 
funds are allocated to states and school districts with greater numbers 
and percentages of children from low-income families. First, using a 
less restrictive hold-harmless provision would reduce the variation in 
funding among school districts with similar numbers and similar 
percentages of poor children and allocate more funding to states with 
more rapidly growing numbers of poor children. Second, funding targeted 
grants and raising the basic grant eligibility thresholds would each 
shift funding toward districts with higher percentages of poor children 
and away from districts with lower percentages of poor children. Third, 
replacing the measure of state per-pupil expenditures with an 
alternative cost indicator would also have the effect of shifting 
funding to districts with higher percentages of poor children and 
reducing the variation in funding among districts with similar 
percentages of poor children. The effect of such changes would depend 
on the appropriations provisions. The effect of the changes would also 
depend on adjustments to the formula-calculated amounts made by state 
officials. Each policy option involves trade-offs between the competing 
goals of providing similar funding amounts to districts with similar 
numbers and percentages of poor children and providing stable funding 
to districts with rapidly declining numbers and percentages of poor 
children. Without increases in total funding, each change also would 
increase funds for some districts while decreasing funds for others. 

Using Less Restrictive Hold-Harmless Rules Would Decrease the Variation 
in Funding Among Districts With Similar Numbers and Percentages of Poor 
Children: 

Different hold-harmless rules affect the extent to which Title I 
funding is allocated on the basis of numbers of poor children. The less 
restrictive the hold-harmless rules are, all else equal, the more Title 
I funding would be allocated on the basis of numbers of poor children. 
We considered four possible hold-harmless rules. In order from most to 
least restrictive, they were: 

* a 100-percent hold-harmless rule (districts receive 100 percent of 
their previous year’s allocation); 

* the school year 2001-02 hold-harmless rule (districts guaranteed the 
larger of either their previous year’s allocation or the amount they 
would have received had the authorizing statute’s hold-harmless rule 
been in effect); 

* the authorizing statute’s hold-harmless rule (districts receive 85, 
90, or 95 percent of the previous year’s allocation, depending on the 
percentage of children in the district who are eligible under the Title 
I formula); and: 

* no hold-harmless rule (districts receive allocations based on current 
application of the Title I formula with no regard to the previous 
year’s allocation). 

Using a less restrictive hold-harmless rule would not noticeably 
redistribute funding between school districts with large and small 
numbers of poor (formula) children. However, less restrictive hold-
harmless rules would substantially reduce the funding variation among 
the smallest districts with similar numbers of poor children. 
Disparities in funding are greatest under the 100-percent hold-harmless 
rule, less under the 2001-02 rule, further reduced under the 
authorizing statute and would be the least if there were no hold-
harmless, as shown in table 5 for both small and large districts, in 
terms of their numbers of formula children.[Footnote 35] 

Table 5: Percentage Differences in Average Funding Variation Under 
Alternative Hold-Harmless Rules Compared With a 100-Percent Rule: 

Dollars per poor child: 

Hold-harmless rule: 100-Percent rule: Average funding; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): $841; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): $814. 

Hold-harmless rule: 100-Percent rule: Average variation; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): $277; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): $213. 

Hold-harmless rule: 2001-02 rule: Average variation; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): $270; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): $209. 

Hold-harmless rule: 2001-02 rule: Percentage difference; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): -2%; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): -2%. 

Hold-harmless rule: Authorizing statute: Average variation; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): $172; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): $205. 

Hold-harmless rule: Authorizing statute: Percentage difference; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): -38%; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): -4%. 
 
Hold-harmless rule: No hold-harmless: Average variation; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): $130; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): $202. 

Hold-harmless rule: No hold-harmless: Percentage difference; 
Smallest districts (the 10,929 districts with up to 643 formula 
children each): -53%; 
Largest districts (the 29 districts with more than 26,976 formula 
children each): -5%. 

Source: GAO Analysis. 

Using a less restrictive hold-harmless rule would also have the effect 
of reducing variation among districts with similar percentages of poor 
children. (See fig. 5.) 

Figure 5: Variation in Funding Per Child Under Alternative Hold-
Harmless Rules (Districts Grouped by Percentage of Poor Children): 

[Refer to PDF for image] 

This figure is a multiple line graph. The vertical axis of the graph 
represents percent variation. The horizontal axis of the graph 
represents children grouped by percentage of poor children in their 
school districts, from lowest to highest. The following alternatives 
are depicted: 
FY 2001 hold-harmless; 
100-Percent hold-harmless; 
Authorizing hold-harmless; 
No hold-harmless. 

Source: GAO Analysis. 

[End of figure] 

Using a less restrictive hold-harmless provision distributes more funds 
to high-growth school districts and accordingly to high-growth states. 
The effect of using a less restrictive hold-harmless rule would be to 
increase the responsiveness of Title I funding to the growth in numbers 
of poor children in states. In figure 6, states are rank-ordered based 
on the growth in the number of formula-eligible children between school 
years 2000-01 and 2001-02.[Footnote 36] Figure 6 shows the percent 
change in funding that would have resulted had no hold-harmless rule 
been in effect. Under the no hold-harmless scenario, 12 of the 15 
states with the highest growth in low-income children would have 
received more funding and 13 of the 15 slowest growth states would have 
received less.[Footnote 37] Appendix II shows the data on which figure 
6 is based. 

Figure 6: Percentage Change in Funding With No Hold-Harmless Rule 
(States Ranked by Percentage Growth in the Number of Poor Children): 

[Refer to PDF for image] 

This figure is a horizontal bar graph. The vertical axis of the graph 
represents percent changes in state poverty from -19.9 to +39.6. The 
horizontal axis of the graph represents percent change in funding from -
30.0 to +15.0. 

Source: GAO analysis. 

[End of figure] 

Distributing Funds Through the Targeted Grants Would Increase the 
Extent to Which Title I Funds Are Allocated to Areas With High Numbers 
and Percentages of Poor Children: 

Funding targeted grants instead of concentration grants would provide 
noticeably more Title I funds to districts with both higher numbers and 
percentages of poor children and reduce funding for districts with 
lower numbers and percentages of poor children. Districts are eligible 
for concentration grants if they have more than 6,500 poor children or 
the number of poor children is more than 15 percent of the district’s 
school-age children. Concentration grants are allocated to eligible 
districts based on their numbers of poor children. In contrast, 
districts would be eligible for targeted grants if they had at least 10 
poor children and these children accounted for at least 5 percent of 
their school-age children. As the number and percentage of poor 
children in the district increase, the targeted grant amount would 
increase both in absolute dollars and proportionally to other 
districts. 

In the 2001-02 school year, districts with the highest percentages of 
poor children received $864 per poor child compared with $758 per poor 
child in districts with the lowest percentages of poor children. If 
targeted grants had been funded instead of concentration grants, the 
funding for districts with the highest percentages of poor children 
would have increased by 5.5 percent, to $912 per poor child, while the 
funding for districts with the lowest percentages of poor children 
would have decreased by 5.7 percent, to $714, as shown in table 6. 
Table 6 also shows that funding targeted grants would have a similar 
effect on districts with larger and smaller numbers of poor children. 

Table 6: Funding Per Poor Child Under Current Law and Under a Policy 
Option That Funds Targeted Grants Instead of Concentration Grants: 

Poverty group by percentage of children: 

Formula option: 2001 formula-calculated amount: Dollars per poor child; 
Lowest (no more than 15% poverty): $758; 
Low (15-21% poverty): $821; 
Medium (21-29% poverty): $826; 
High (29-40% poverty): $865; 
Highest (40-100% poverty): $864; 
Ratio (highest to lowest): 114%. 

Formula option: Formula-calculated amount with targeted grant: Dollars 
per poor child; 
Lowest (no more than 15% poverty): $714; 
Low (15-21% poverty): $808; 
Medium (21-29% poverty): $820; 
High (29-40% poverty): $877; 
Highest (40-100% poverty): $912; 
Ratio (highest to lowest): 128%. 

Formula option: Formula-calculated amount with targeted grant: 
Percentage difference; 
Lowest (no more than 15% poverty): -5.7%; 
Low (15-21% poverty): -1.6%; 
Medium (21-29% poverty): -0.7%; 
High (29-40% poverty): +1.4%; 
Highest (40-100% poverty): +5.5%. 

Poverty group by numbers of children in poverty: 

Formula option: 2001 formula-calculated amount: Dollars per poor child; 
Lowest (1 to 643 formula children): $842; 
Low (643-2,020 formula children): $797; 
Medium (2.021 to 6,698 formula children): $784; 
High (6,709 to 26,844 formula children): $821; 
Highest (26,976 to 303,122 formula children): $814; 
Ratio to lowest): 97%. 

Formula option: Formula-calculated amount with targeted grant: Dollars 
per poor child; 
Lowest (1 to 643 formula children): $815; 
Low (643-2,020 formula children): $785; 
Medium (2.021 to 6,698 formula children): $771; 
High (6,709 to 26,844 formula children): $812; 
Highest (26,976 to 303,122 formula children): $946; 
Ratio to lowest): 116%. 

Formula option: Formula-calculated amount with targeted grant: 
Percentage difference; 
Lowest (1 to 643 formula children): -3.1%; 
Low (643-2,020 formula children): -1.6%; 
Medium (2.021 to 6,698 formula children): -1.6%; 
High (6,709 to 26,844 formula children): -1.1%; 
Highest (26,976 to 303,122 formula children): +16.3%. 

Source: GAO Analysis. 

[End of table] 

Alternative Cost Indicator Would Direct More Funding to Districts With 
Higher Percentages of Poor Children and Reduce Funding Disparities 
Among These Districts: 

The per-pupil expenditure factor was originally included in the Title I 
formulas to take into account cross-state differences in the cost of 
providing education services. While per-pupil expenditures reflect the 
cost of providing education services to some extent, expenditures are 
also explained by other factors not related to costs. For example, 
states with high-income taxpayers may spend more on education than 
those whose taxpayers have lower incomes.[Footnote 38] In addition, 
spending differences may result from differences in the “willingness” 
of a state’s taxpayers to fund public education.[Footnote 39] 

One alternative cost measure is a geographical cost-of-education index 
developed for the Department of Education’s Office of Educational 
Research and Improvement.[Footnote 40] The purpose of this experimental 
cost index is to make cost comparisons based on the cost of teachers, 
non-teaching school personnel, and other factors that may affect costs, 
but which are beyond the ability of local officials to control. 
[Footnote 41] The index includes cost factors for both states and 
school districts, unlike earlier experimental measures that had only 
cross-state cost factors. We use these cost factors for illustrative 
purposes only and do not necessarily endorse any particular measure. 

The intent of replacing state per-pupil expenditures with either a 
state or district-level cost-of-education factor is to more accurately 
reflect educational costs; however, as a by-product, doing so would 
shift funding somewhat toward districts with higher percentages of poor 
children. If the state per-pupil expenditure factor had been replaced 
with a cost factor in the 2001-02 school year, districts with the 
highest percentages of poor children would have seen an increase in 
funding of approximately 3 percent, while districts with the lowest 
percentages of poor children would have seen a decrease in funding of 2 
to 3 percent (see table 7). If in addition, the authorizing statute’s 
hold-harmless rules had been adopted, funding to districts with the 
highest percentages of poor children would have increased by about 5 
percent while funding in districts with the lowest percentages of poor 
children would have decreased by about 5 percent.[Footnote 42] 

Table 7: Funding Per Poor Child Under Current Law and Formula Options 
That Replace State Per-Pupil Expenditure Factor With a Teacher Cost 
Factor: 

Poverty group by percentage of children in poverty: 

2001 formula allocation: Dollars per child; 
Lowest (no more than 15% poverty): $758; 
Low (15-21% poverty): $821; 
Medium (21-29% poverty): $826; 
High (29-40% poverty): $865; 
Highest (40-100% poverty): $864; 
Ratio (highest to lowest): 114%. 

State cost factor: Dollars per child; 
Lowest (no more than 15% poverty): $737; 
Low (15-21% poverty): $815; 
Medium (21-29% poverty): $825; 
High (29-40% poverty): $866; 
Highest (40-100% poverty): $892; 
Ratio (highest to lowest): 121%. 

State cost factor: Percentage difference; 
Lowest (no more than 15% poverty): -2.7%; 
Low (15-21% poverty): -0.8%; 
Medium (21-29% poverty): 0.0; 
High (29-40% poverty): 0.1%; 
Highest (40-100% poverty): 3.3%. 

District cost factor: Dollars per child; 
Lowest (no more than 15% poverty): $746; 
Low (15-21% poverty): $815; 
Medium (21-29% poverty): $821; 
High (29-40% poverty): $866; 
Highest (40-100% poverty): $887; 
Ratio (highest to lowest): 119%. 

District cost factor: Percentage difference; 
Lowest (no more than 15% poverty): -1.6%; 
Low (15-21% poverty): -0.8%; 
Medium (21-29% poverty): -0.5%; 
High (29-40% poverty): 0.1%; 
Highest (40-100% poverty): 2.7%. 

District cost factor & authorizing hold-harmless: Dollars per child; 
Lowest (no more than 15% poverty): $716; 
Low (15-21% poverty): $818; 
Medium (21-29% poverty): $818; 
High (29-40% poverty): $873; 
Highest (40-100% poverty): $905; 
Ratio (highest to lowest): 126%. 

District cost factor & authorizing hold-harmless: Percentage 
difference; 
Lowest (no more than 15% poverty): -5.4%; 
Low (15-21% poverty): -0.5%; 
Medium (21-29% poverty): -1.0%; 
High (29-40% poverty): 0.9%; 
Highest (40-100% poverty): 4.7%. 

Source: GAO Analysis. 

The more substantial effect of replacing the state per-pupil 
expenditure factor, however, would be to reduce the variation in 
funding among districts with higher percentages of poor children (see 
figure 7). With the educational cost factor, the variation in funding 
among districts with higher percentages of poor children would be 
reduced from 19 percent to 13 percent, a reduction of 30 percent. In 
contrast, the variation in funding among districts with low percentages 
of poor children would be largely unaffected. However, if the current 
hold-harmless rules were also replaced with the authorizing statutes 
rules, very substantial reductions in funding disparities would result 
among all school districts. Funding variations would be cut by more 
than half between the highest and lowest poverty districts. 

Figure 7: Comparison of the Variation in Funding Per Child Across 
School Districts Under Formula Options That Replace State Per-Pupil 
Spending With a Cost of Education Factor (Fiscal Year 2001): 

[Refer to PDF for image] 

This figure is a multiple line graph. The vertical axis of the graph 
represents percent coefficient of variation. The horizontal axis of the 
graph represents children grouped by percentage of poor children in 
their school districts, from lowest to highest. The following funding 
factors are depicted: 
2001 Formula allocation; 
State cost factor; 
District cost factor; 
District cost factor/authorizing hold-harmless. 

Source: GAO Analysis. 

[End of figure] 

Raising the Poverty Threshold for Eligibility to Qualify for Basic 
Grants Would Potentially Affect Large Numbers of Districts With Low 
Percentages of Poor Children: 

Title I grants have sometimes been criticized because the poverty 
threshold for basic grant eligibility is so low and that nearly all 
districts can participate in the program.[Footnote 43] It is often 
noted that by funding nearly all districts, less funding is available 
for high-poverty districts. One policy option is to raise the basic 
grant eligibility threshold, making fewer districts eligible. With 
fewer districts eligible, the remaining districts would receive more 
funds per poor child, if total funding were to remain constant. Table 8 
shows how increasing the current 2 percent poverty threshold to poverty 
thresholds of 5 and 10 percent, respectively, would have this effect. 
For example, a 10 percent threshold would result in 26 percent of all 
districts, which contain 7.7 percent of all formula children, becoming 
ineligible. This would allow funding per child in the remaining 
districts to increase by 8.3 percent or $57. The effects of these 
thresholds are shown on a state-by-state basis in table 9. 

Table 8: Number of Eligible School Districts, Children, and Funding Per 
Child Under Alternative Eligibility Thresholds: 

Alternative Basic-Grant Eligibility Threshold: 

Number of ineligible school districts (below eligibility threshold): 
Current basic grant 2% eligibility threshold: 168; 
5% Eligibility threshold: 1,148; 
10% Eligibility threshold: 3,527. 

Percentage of all school districts: 
Current basic grant 2% eligibility threshold: 1.3%; 
5% Eligibility threshold: 8.6%; 
10% Eligibility threshold: 26.3%. 

Number of children in ineligible districts (district doesn’t meet 
eligibility threshold): 
Current basic grant 2% eligibility threshold: 7,141; 
5% Eligibility threshold: 130,144; 
10% Eligibility threshold: 779,684. 

Percentage of all formula children: 
Current basic grant 2% eligibility threshold: 0.07%; 
5% Eligibility threshold: 1.3%; 
10% Eligibility threshold: 7.7%. 

Eligible district’s average school year 2001 basic allotment per 
eligible formula child: 
Current basic grant 2% eligibility threshold: $684; 
5% Eligibility threshold: $693; 
10% Eligibility threshold: $741. 

Percentage increase in funding to remaining eligible districts resulting
from increase in eligibility threshold: 
Current basic grant 2% eligibility threshold: NA; 
5% Eligibility threshold: 1.3%; 
10% Eligibility threshold: 8.3%. 

Note: Numbers and percentages exclude districts that would be 
ineligible because they have fewer than 10 formula children. 

Source: GAO Analysis. 

[End of table] 

Table 9: Number of Ineligible Districts, Poor Children, and Title I 
Funding They Would Receive Under Alternative Eligibility Thresholds: 

State: Alabama; 
5% Eligibility threshold: Number of districts made ineligible: 6; 
5% Eligibility threshold: Children in ineligible districts: 490; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $368,973
10% Eligibility threshold: Number of districts made ineligible: 8; 
10% Eligibility threshold: Children in ineligible districts: 4,458
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $3,729,223. 

State: Alaska; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 7; 
10% Eligibility threshold: Children in ineligible districts: 940; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $776,674. 

State: Arizona; 
5% Eligibility threshold: Number of districts made ineligible: 7; 
5% Eligibility threshold: Children in ineligible districts: 443; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $324,165; 
10% Eligibility threshold: Number of districts made ineligible: 20; 
10% Eligibility threshold: Children in ineligible districts: 14,348; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $7,851,414. 

State: Arkansas; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 5; 
10% Eligibility threshold: Children in ineligible districts: 984; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $675,864. 

State: California; 
5% Eligibility threshold: Number of districts made ineligible: 46; 
5% Eligibility threshold: Children in ineligible districts: 6,687
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $4,000,328
10% Eligibility threshold: Number of districts made ineligible: 166; 
10% Eligibility threshold: Children in ineligible districts: 55,800
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $34,926,296. 

State: Colorado; 
5% Eligibility threshold: Number of districts made ineligible: 5; 
5% Eligibility threshold: Children in ineligible districts: 1,610; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $967,747; 
10% Eligibility threshold: Number of districts made ineligible: 33; 
10% Eligibility threshold: Children in ineligible districts: 23,044; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $13,893,702. 

State: Connecticut; 
5% Eligibility threshold: Number of districts made ineligible: 59; 
5% Eligibility threshold: Children in ineligible districts: 4,871; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $3,924,895; 
10% Eligibility threshold: Number of districts made ineligible: 116; 
10% Eligibility threshold: Children in ineligible districts: 16,412; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $13,273,760. 

State: Delaware; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 1; 
10% Eligibility threshold: Children in ineligible districts: 224; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $203,485. 

State: District of Columbia; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Florida; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Georgia; 
5% Eligibility threshold: Number of districts made ineligible: 1; 
5% Eligibility threshold: Children in ineligible districts: 23; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $14,542; 
10% Eligibility threshold: Number of districts made ineligible: 6; 
10% Eligibility threshold: Children in ineligible districts: 20,004; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $12,648,100. 

State: Hawaii; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Idaho; 
5% Eligibility threshold: Number of districts made ineligible: 2; 
5% Eligibility threshold: Children in ineligible districts: 64; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $57,553; 
10% Eligibility threshold: Number of districts made ineligible: 10; 
10% Eligibility threshold: Children in ineligible districts: 2,300; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $1,494,645. 

State: Illinois; 
5% Eligibility threshold: Number of districts made ineligible: 171; 
5% Eligibility threshold: Children in ineligible districts: 12,908; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $9,831,050; 
10% Eligibility threshold: Number of districts made ineligible: 386; 
10% Eligibility threshold: Children in ineligible districts: 44,267; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $33,665,002. 

State: Indiana; 
5% Eligibility threshold: Number of districts made ineligible: 28; 
5% Eligibility threshold: Children in ineligible districts: 3,260; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $2,400,094; 
10% Eligibility threshold: Number of districts made ineligible: 128; 
10% Eligibility threshold: Children in ineligible districts: 25,197; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $17,963,907. 

State: Iowa; 
5% Eligibility threshold: Number of districts made ineligible: 20; 
5% Eligibility threshold: Children in ineligible districts: 1,148; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $1,020,095; 
10% Eligibility threshold: Number of districts made ineligible: 128; 
10% Eligibility threshold: Children in ineligible districts: 10,395; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $8,267,730. 

State: Kansas; 
5% Eligibility threshold: Number of districts made ineligible: 9; 
5% Eligibility threshold: Children in ineligible districts: 760; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $552,810; 
10% Eligibility threshold: Number of districts made ineligible: 63; 
10% Eligibility threshold: Children in ineligible districts: 9,411; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $6,437,589. 

State: Kentucky; 
5% Eligibility threshold: Number of districts made ineligible: 2; 
5% Eligibility threshold: Children in ineligible districts: 285; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $174,466; 
10% Eligibility threshold: Number of districts made ineligible: 8; 
10% Eligibility threshold: Children in ineligible districts: 3,431; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $2,145,994. 

State: Louisiana; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Maine; 
5% Eligibility threshold: Number of districts made ineligible: 12; 
5% Eligibility threshold: Children in ineligible districts: 467; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $620,171; 
10% Eligibility threshold: Number of districts made ineligible: 51; 
10% Eligibility threshold: Children in ineligible districts: 3,643; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $3,936,535. 

State: Maryland; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 7; 
10% Eligibility threshold: Children in ineligible districts: 33,679; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $25,483,146. 

State: Massachusetts; 
5% Eligibility threshold: Number of districts made ineligible: 77; 
5% Eligibility threshold: Children in ineligible districts: 6,068
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $5,385,737; 
10% Eligibility threshold: Number of districts made ineligible: 177; 
10% Eligibility threshold: Children in ineligible districts: 24,133; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $20,302,292. 

State: Michigan; 
5% Eligibility threshold: Number of districts made ineligible: 52; 
5% Eligibility threshold: Children in ineligible districts: 10,299; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $9,577,340; 
10% Eligibility threshold: Number of districts made ineligible: 161; 
10% Eligibility threshold: Children in ineligible districts: 40,335; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $36,122,572. 

State: Minnesota; 
5% Eligibility threshold: Number of districts made ineligible: 20; 
5% Eligibility threshold: Children in ineligible districts: 6,257; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $4,299,630; 
10% Eligibility threshold: Number of districts made ineligible: 103; 
10% Eligibility threshold: Children in ineligible districts: 31,867; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $22,062,822. 

State: Mississippi; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Missouri; 
5% Eligibility threshold: Number of districts made ineligible: 16; 
5% Eligibility threshold: Children in ineligible districts: 4,840; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $3,295,938; 
10% Eligibility threshold: Number of districts made ineligible: 75; 
10% Eligibility threshold: Children in ineligible districts: 20,267; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $14,583,945. 

State: Montana; 
5% Eligibility threshold: Number of districts made ineligible: 1; 
5% Eligibility threshold: Children in ineligible districts: 23; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $14,289; 
10% Eligibility threshold: Number of districts made ineligible: 29; 
10% Eligibility threshold: Children in ineligible districts: 661; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $522,560. 

State: Nebraska; 
5% Eligibility threshold: Number of districts made ineligible: 14; 
5% Eligibility threshold: Children in ineligible districts: 1,884; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $1,397,013; 
10% Eligibility threshold: Number of districts made ineligible: 100; 
10% Eligibility threshold: Children in ineligible districts: 6,788; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $5,642,142. 

State: Nevada; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 2; 
10% Eligibility threshold: Children in ineligible districts: 1,074; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $608,365. 

State: New Hampshire; 
5% Eligibility threshold: Number of districts made ineligible: 27; 
5% Eligibility threshold: Children in ineligible districts: 1,618; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $1,459,650; 
10% Eligibility threshold: Number of districts made ineligible: 84; 
10% Eligibility threshold: Children in ineligible districts: 8,265; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $7,315,072. 

State: New Jersey; 
5% Eligibility threshold: Number of districts made ineligible: 169; 
5% Eligibility threshold: Children in ineligible districts: 12,890; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $11,265,346; 
10% Eligibility threshold: Number of districts made ineligible: 346; 
10% Eligibility threshold: Children in ineligible districts: 41,109; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $35,640,073. 

State: New Mexico; 
5% Eligibility threshold: Number of districts made ineligible: 2; 
5% Eligibility threshold: Children in ineligible districts: 494; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $316,533; 
10% Eligibility threshold: Number of districts made ineligible: 3; 
10% Eligibility threshold: Children in ineligible districts: 541; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $341,780. 

State: New York; 
5% Eligibility threshold: Number of districts made ineligible: 77; 
5% Eligibility threshold: Children in ineligible districts: 9,068; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $7,796,900; 
10% Eligibility threshold: Number of districts made ineligible: 223; 
10% Eligibility threshold: Children in ineligible districts: 48,460; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $39,926,375. 

State: North Carolina; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 1; 
10% Eligibility threshold: Children in ineligible districts: 1,788; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $1,047,215. 

State: North Dakota; 
5% Eligibility threshold: Number of districts made ineligible: 2; 
5% Eligibility threshold: Children in ineligible districts: 17; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $14,578; 
10% Eligibility threshold: Number of districts made ineligible: 21; 
10% Eligibility threshold: Children in ineligible districts: 1,099; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $1,018,412. 

State: Ohio; 
5% Eligibility threshold: Number of districts made ineligible: 107; 
5% Eligibility threshold: Children in ineligible districts: 12,624; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $10,803,019; 
10% Eligibility threshold: Number of districts made ineligible: 294; 
10% Eligibility threshold: Children in ineligible districts: 49,332; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $40,399,218. 

State: Oklahoma; 
5% Eligibility threshold: Number of districts made ineligible: 5; 
5% Eligibility threshold: Children in ineligible districts: 366; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $295,932; 
10% Eligibility threshold: Number of districts made ineligible: 39; 
10% Eligibility threshold: Children in ineligible districts: 6,841; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $4,554,951. 

State: Oregon; 
5% Eligibility threshold: Number of districts made ineligible: 4; 
5% Eligibility threshold: Children in ineligible districts: 580; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $449,338; 
10% Eligibility threshold: Number of districts made ineligible: 26; 
10% Eligibility threshold: Children in ineligible districts: 10,429; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $7,873,530. 

State: Pennsylvania; 
5% Eligibility threshold: Number of districts made ineligible: 61; 
5% Eligibility threshold: Children in ineligible districts: 9,522
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $7,774,477; 
10% Eligibility threshold: Number of districts made ineligible: 176; 
10% Eligibility threshold: Children in ineligible districts: 46,438; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $37,347,170. 

State: Puerto Rico; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Rhode Island; 
5% Eligibility threshold: Number of districts made ineligible: 6; 
5% Eligibility threshold: Children in ineligible districts: 582; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $468,957; 
10% Eligibility threshold: Number of districts made ineligible: 23; 
10% Eligibility threshold: Children in ineligible districts: 6,187; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $4,985,285. 

State: South Carolina; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0;; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 3; 
10% Eligibility threshold: Children in ineligible districts: 1,380; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $776,247. 

State: South Dakota; 
5% Eligibility threshold: Number of districts made ineligible: 3; 
5% Eligibility threshold: Children in ineligible districts: 111; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $100,455; 
10% Eligibility threshold: Number of districts made ineligible: 20; 
10% Eligibility threshold: Children in ineligible districts: 926; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $775,148. 

State: Tennessee; 
5% Eligibility threshold: Number of districts made ineligible: 2; 
5% Eligibility threshold: Children in ineligible districts: 1,023; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $644,205; 
10% Eligibility threshold: Number of districts made ineligible: 7; 
10% Eligibility threshold: Children in ineligible districts: 6,622; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $4,713,538. 

State: Texas; 
5% Eligibility threshold: Number of districts made ineligible: 31; 
5% Eligibility threshold: Children in ineligible districts: 11,355; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $7,457,363; 
10% Eligibility threshold: Number of districts made ineligible: 110; 
10% Eligibility threshold: Children in ineligible districts: 51,933; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $32,388,180. 

State: Utah; 
5% Eligibility threshold: Number of districts made ineligible: 1; 
5% Eligibility threshold: Children in ineligible districts: 111; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $67,570; 
10% Eligibility threshold: Number of districts made ineligible: 10; 
10% Eligibility threshold: Children in ineligible districts: 17,050; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $9,601,315. 

State: Vermont; 
5% Eligibility threshold: Number of districts made ineligible: 20; 
5% Eligibility threshold: Children in ineligible districts: 701; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $886,384; 
10% Eligibility threshold: Number of districts made ineligible: 72; 
10% Eligibility threshold: Children in ineligible districts: 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $3,597,152. 

State: Virginia; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 17; 
10% Eligibility threshold: Children in ineligible districts: 29,989; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $17,618,948. 

State: Washington; 
5% Eligibility threshold: Number of districts made ineligible: 8; 
5% Eligibility threshold: Children in ineligible districts: 1,641; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $1,052,712; 
10% Eligibility threshold: Number of districts made ineligible: 52; 
10% Eligibility threshold: Children in ineligible districts: 26,901; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $16,951,548. 

State: West Virginia; 
5% Eligibility threshold: Number of districts made ineligible: 0; 
5% Eligibility threshold: Children in ineligible districts: 0; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: 0; 
10% Eligibility threshold: Number of districts made ineligible: 0; 
10% Eligibility threshold: Children in ineligible districts: 0; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: 0. 

State: Wisconsin; 
5% Eligibility threshold: Number of districts made ineligible: 74; 
5% Eligibility threshold: Children in ineligible districts: 5,018; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $4,224,246; 
10% Eligibility threshold: Number of districts made ineligible: 202; 
10% Eligibility threshold: Children in ineligible districts: 26,036; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $21,483,181. 

State: Wyoming; 
5% Eligibility threshold: Number of districts made ineligible: 1; 
5% Eligibility threshold: Children in ineligible districts: 36; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $111,155; 
10% Eligibility threshold: Number of districts made ineligible: 8; 
10% Eligibility threshold: Children in ineligible districts: 1,921; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $2,221,161. 

Total: 
5% Eligibility threshold: Number of districts made ineligible: 1,148; 
5% Eligibility threshold: Children in ineligible districts: 130,144; 
5% Eligibility threshold: Total FY 2001 basic grant dollars reallocated 
from ineligible districts: $103,415,654; 
10% Eligibility threshold: Number of districts made ineligible: 3,527; 
10% Eligibility threshold: Children in ineligible districts: 779,684; 
10% Eligibility threshold: Total FY 2001 basic grant dollars 
reallocated from ineligible districts: $577,793,262. 

Source: GAO Analysis. 

Most States Alter Formula-Calculated Amounts When Allocating Funds to 
Their School Districts: 

Changes to the Title I allocation formulas will change the amount of 
funds states receive and also would be expected to result in changes in 
the amounts districts receive; however, there are limits on how 
precisely changes in the formula can be expected to affect school 
districts because states alter the formula-calculated amounts. In the 
aggregate, relatively few poor children and Title I funds were 
associated with districts whose allocations differed widely from their 
formula-calculated amounts in the 1999-2000 school year. As a result, 
state adjustments did not appear to alter the overall extent to which 
available funding was allocated on the basis of the number of poor 
children. However, for some individual states and school districts, 
state adjustments were substantial. 

When allocating the funds they receive from the federal government, 
states adjust for changes in school district boundaries and the 
creation of charter schools. In our work, we found that among the 
school districts operating in the 1999-2000 school year, there were 
more than 900 school districts, containing about 126,000 children 
eligible for free or reduced-price lunch and receiving about $79 
million in Title I funds, that were not included in Education’s formula 
calculations. In addition, states alter the formula-calculated amounts 
to adjust for the poverty measure used for school districts in small 
areas, as well as to fund statewide activities and program 
administration. States are allowed to withhold up to 1.5 percent of 
their Title I funds for statewide activities and program 
administration, so this much variation from the formula-calculated 
allocations is expected. 

Considering only the districts that were included in Education’s 
calculations, we found that during the 1999-2000 school year 49 percent 
of the districts received total Title I grants that differed by more 
than 1.5 percent from the formula-calculated allocation; 16 percent of 
districts’ allocations differed by more than 10 percent. Among the 
districts included in Education’s calculations, over half of the Title 
I funds were allocated to districts whose allocation amounts differed 
from their formula calculations by no more than 1.5 percent. These 
districts also contained just over half of the poor children. Only 
about 5 percent of the funds were allocated to districts whose actual 
allocations differed from their formula calculations by more than 10 
percent. These districts also contained about 5 percent of the poor 
children. (See table 10.) 

Table 10: Distribution of Districts, Poor Children, and Title I Dollars 
by Extent of Difference Between District Allocations and Formula 
Calculations: 

Number of school districts: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 7,009; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 4,554; 
Districts with allocations differing from formula calculation by more 
than 10%: 2,206; 
Total: 13,769. 

Percentage of school districts: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 51%; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 33%; 
Districts with allocations differing from formula calculation by more 
than 10%: 16%; 
Total: 100%. 

Number of Census poor children: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 5,065,175; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 3,644,038; 
Districts with allocations differing from formula calculation by more 
than 10%: 505,162; 
Total: 9,214,375. 

Percentage of Census poor children: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 55%; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 40%; 
Districts with allocations differing from formula calculation by more 
than 10%: 5%; 
Total: 100%. 

Number of children eligible for free or reduced-price lunch: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 9,485,789; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 7,187,813; 
Districts with allocations differing from formula calculation by more 
than 10%: 910,791; 
Total: 17,584,393. 

Percentage of children eligible for free or reduced-price lunch: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 54%; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 41%; 
Districts with allocations differing from formula calculation by more 
than 10%: 5%; 
Total: 100%. 

Formula-calculated dollar amounts: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: $4,150,777,359; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: $2,702,779,708; 
Districts with allocations differing from formula calculation by more 
than 10%: $401,904,928; 
Total: $7,255,461,995. 

Percentage of formula-calculated dollars: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 57%; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 37%; 
Districts with allocations differing from formula calculation by more 
than 10%: 6%; 
Total: 100%. 

Actual allocation dollar amounts: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: $4,129,794,331; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: $2,630,272,381; 
Districts with allocations differing from formula calculation by more 
than 10%: $348,916,755; 
Total: $7,108,983,467. 

Percentage of actual allocations: 
Districts with allocations differing from formula calculation by no 
more than 1.5%: 58%; 
Districts with allocations differing from formula calculation by 
between 1.5% and 10%: 37%; 
Districts with allocations differing from formula calculation by more 
than 10%: 5%; 
Total: 100%. 

Source: GAO Analysis. 

[End of table] 

The variation between actual allocations and the formula calculations 
is greater in some states than in others. There are some states, for 
example, Georgia, Louisiana, and South Dakota, where actual allocations 
to school districts are very close to the formula calculations. (See 
table 11.) 

Table 11: Percentage of Poor Children and Percentage of Dollars 
Allocated to Districts Whose Allocations Differed From Formula 
Calculations by More Than 1.5 Percent: 

Percentage of districts with allocations differing from formula 
calculations by more than 1.5 percent: 
South Dakota: 6%; 
Georgia: 18%; 
Louisiana: 16%. 

Percentage of Census poor children in districts with allocations 
differing from formula calculations by more than 1.5 percent: 
South Dakota: 6%; 
Georgia: 29%; 
Louisiana: 46%. 


Percentage of Title I allocations received in districts with 
allocations differing from formula calculations by more than 1.5 
percent: 
South Dakota: 6%; 
Georgia: 28%; 
Louisiana: 40%. 

Source: GAO Analysis. 

[End of table] 

However, these states are the exception, as most states, including 
Delaware, North Dakota, and Maine, had many districts receiving very 
different allocations than the formula calculations. (See table 12.) 

Table 12: Percentage of Poor Children and Percentage of Dollars 
Allocated to Districts Whose Allocations Differed From Formula 
Calculations by More Than 10 Percent: 

Percentage of districts with allocations differing from formula 
calculations by more than 10 percent: 
North Dakota: 62%; 
Delaware: 49%; 
Maine: 40%. 

Percentage of Census poor children in districts with allocations 
differing from formula calculations by more than 10 percent: 
North Dakota: 51%; 
Delaware: 41%; 
Maine: 31%. 

Percentage of Title I allocations received in districts with 
allocations differing from formula calculations by more than 10 
percent: 
North Dakota: 56%; 
Delaware: 41%; 
Maine: 29%. 

Source: GAO Analysis. 

[End of table] 

However, states do not appear to alter the overall extent to which 
available funding is allocated on the basis of the number of poor 
children. Table 13 shows for each state the percentage of its districts 
whose total Title I grants differed from their formula calculations by 
less than 1.5 percent, between 1.5 and 10 percent, and by more than 10 
percent. 

Table 13: Differences Between Actual Allocations and Formula Calculated 
Amounts, 1999-2000 School Year: 
 
State: Alabama; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 61.1; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 19.8; 
Percentage of districts with grants more than 10% from formula 
calculation: 19.1; 
Total: 100%. 

State: Alaska; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 28.3; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 34.0; 
Percentage of districts with grants more than 10% from formula 
calculation: 37.7; 
Total: 100%. 

State: Arizona; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 11.8; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 52.9; 
Percentage of districts with grants more than 10% from formula 
calculation: 35.3; 
Total: 100%. 

State: Arkansas; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 68.4; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 31.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.0; 
Total: 100%. 

State: California; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 33.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 64.1; 
Percentage of districts with grants more than 10% from formula 
calculation: 2.4; 
Total: 100%. 

State: Colorado; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 71.6; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 19.3; 
Percentage of districts with grants more than 10% from formula 
calculation: 9.1; 
Total: 100%. 

State: Connecticut; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 41.6; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 27.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 30.7; 
Total: 100%. 

State: Delaware; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 6.3; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 31.3; 
Percentage of districts with grants more than 10% from formula 
calculation: 62.4; 
Total: 100%. 

State: Florida; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 40.3; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 59.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.0; 
Total: 100%. 

State: Georgia; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 94.4; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 5.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.0; 
Total: 100%. 

State: Idaho; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 71.4; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 37.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.9; 
Total: 100%. 

State: Illinois; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 69.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 14.4; 
Percentage of districts with grants more than 10% from formula 
calculation: 16.1; 
Total: 100%. 

State: Indiana; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 56.7; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 31.8; 
Percentage of districts with grants more than 10% from formula 
calculation: 11.5; 
Total: 100%. 

State: Iowa; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 39.8; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 41.5; 
Percentage of districts with grants more than 10% from formula 
calculation: 18.7; 
Total: 100%. 

State: Kansas; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 55.3; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 27.8; 
Percentage of districts with grants more than 10% from formula 
calculation: 16.9; 
Total: 100%. 

State: Kentucky; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 0.6; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 94.9; 
Percentage of districts with grants more than 10% from formula 
calculation: 4.5; 
Total: 100%. 

State: Louisiana; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 81.8; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 13.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 4.6; 
Total: 100%. 

State: Maine; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 35.0; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 25.1; 
Percentage of districts with grants more than 10% from formula 
calculation: 39.9; 
Total: 100%. 

State: Maryland; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 29.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 70.8; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.0; 
Total: 100%. 

State: Massachusetts; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 43.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 10.0; 
Percentage of districts with grants more than 10% from formula 
calculation: 46.8; 
Total: 100%. 

State: Michigan; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 66.0; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 16.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 17.4; 
Total: 100%. 

State: Minnesota; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 71.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 20.2; 
Percentage of districts with grants more than 10% from formula 
calculation: 8.6; 
Total: 100%. 

State: Mississippi; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 76.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 12.8; 
Percentage of districts with grants more than 10% from formula 
calculation: 10.7; 
Total: 100%. 

State: Missouri; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 30.4; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 29.2; 
Percentage of districts with grants more than 10% from formula 
calculation: 40.4; 
Total: 100%. 

State: Montana; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 70.8; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 18.3; 
Percentage of districts with grants more than 10% from formula 
calculation: 10.9; 
Total: 100%. 

State: Nebraska; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 64.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 13.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 21.8; 
Total: 100%. 

State: Nevada; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 58.8; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 41.2; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.0; 
Total: 100%. 

State: New Hampshire; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 19.8; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 71.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 8.6; 
Total: 100%. 

State: New Jersey; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 31.0; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 22.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 46.4; 
Total: 100%. 

State: New Mexico; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 47.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 46.1; 
Percentage of districts with grants more than 10% from formula 
calculation: 6.7; 
Total: 100%. 

State: New York; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 28.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 70.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.9; 
Total: 100%. 

State: North Carolina; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 43.6; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 54.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 1.7; 
Total: 100%. 

State: North Dakota; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 18.0; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 33.5; 
Percentage of districts with grants more than 10% from formula 
calculation: 48.5; 
Total: 100%. 

State: Ohio; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 76.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 13.4; 
Percentage of districts with grants more than 10% from formula 
calculation: 10.4; 
Total: 100%. 

State: Oklahoma; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 32.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 23.2; 
Percentage of districts with grants more than 10% from formula 
calculation: 44.3; 
Total: 100%. 

State: Oregon; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 61.5; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 21.0; 
Percentage of districts with grants more than 10% from formula 
calculation: 17.5; 
Total: 100%. 

State: Pennsylvania; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 74.3; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 12.3; 
Percentage of districts with grants more than 10% from formula 
calculation: 13.4; 
Total: 100%. 

State: Rhode Island; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 36.1; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 16.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 47.2; 
Total: 100%. 

State: South Carolina; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 44.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 47.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 8.1; 
Total: 100%. 

State: South Dakota; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 84.3; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 8.7; 
Percentage of districts with grants more than 10% from formula 
calculation: 7.0; 
Total: 100%. 

State: Tennessee; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 67.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 24.1; 
Percentage of districts with grants more than 10% from formula 
calculation: 8.7; 
Total: 100%. 

State: Texas; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 56.0; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 43.2; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.8; 
Total: 100%. 

State: Utah; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 70.0; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 17.5; 
Percentage of districts with grants more than 10% from formula 
calculation: 12.5; 
Total: 100%. 

State: Virginia; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 69.9; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 29.3; 
Percentage of districts with grants more than 10% from formula 
calculation: 0.8; 
Total: 100%. 

State: Washington; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 66.9; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 20.6; 
Percentage of districts with grants more than 10% from formula 
calculation: 12.5; 
Total: 100%. 

State: West Virginia; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 69.1; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 21.8; 
Percentage of districts with grants more than 10% from formula 
calculation: 9.1; 
Total: 100%. 

State: Wisconsin; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 6.4; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 89.9;; 
Percentage of districts with grants more than 10% from formula 
calculation: 3.7; 
Total: 100%. 

State: Wyoming; 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 29.2; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 31.3; 
Percentage of districts with grants more than 10% from formula 
calculation: 39.5; 
Total: 100%. 

National: 
Percentage of districts with grants within 1.5% of formula 
calculation[A]: 50.9; 
Percentage of districts with grants between 1.5% and 10% of formula 
calculation: 33.1; 
Percentage of districts with grants more than 10% from formula 
calculation: 16.0; 
Total: 100%. 

Note: Hawaii and District of Columbia excluded, as they are single 
districts. Vermont excluded as Supervisory Unions, not districts, 
utilized for allocating funds. 

[A] Includes those districts calculated by Education to receive $0. 

Source: GAO Analysis. 

[End of table] 

Concluding Observations: 

Although Title I funding generally reflects the distribution of poor 
children, there are many instances of states, districts, and schools 
with either similar numbers or similar percentages of poor children 
receiving widely differing amounts of funding per poor child. These 
differences result, in part, from formula provisions that attempt to 
balance several, sometimes competing, goals. These goals include 
allocating funds based on the distribution of poor children, ensuring 
that states and districts are provided funding stability even in light 
of declining numbers of poor children, and addressing differences 
across school districts and states in the costs of providing 
educational services. Choosing among the policy options discussed in 
this report will entail, in part, weighing the goal of increased 
targeting with other goals. Enacting any of the policy options— using 
less restrictive hold-harmless provisions, funding targeted grants, 
using an alternative cost factor, or raising the eligibility 
threshold—would result in changes for many states and school districts 
in terms of their formula calculations. In addition, under any of these 
policy options, states and school districts would still have 
flexibility in making allocation decisions—flexibility that allows 
states and school districts to use these funds in a manner that they 
believe best meets the needs of disadvantaged children. 

Agency Comments: 

In written comments on our draft report, the Department of Education 
generally agreed with the findings presented in the report. Education 
suggested that our report be updated to reflect the passage of the “No 
Child Left Behind Act of 2001” and the fiscal 2002 appropriations act. 
Appendix III describes the impact of this legislation on the aspects of 
Title I discussed in the report. Education’s written comments are 
printed in appendix IV. 

In written comments on our draft report, the U.S. Department of 
Agriculture (USDA) said that the Food and Nutrition Service (FNS) has 
become increasingly aware of the limitations of free and reduced-price 
lunch data as a measure of low-income status, which could have 
implications for the targeting of Title I funds. As described in the 
report, we recognize that there are limitations of these data as a 
measure of poverty. Despite these limitations, however, we chose to use 
subsidized lunch data as one of our measures of poverty for several 
reasons. We used these data as a poverty measure at the school level 
because the Department of Education has found these data to be the best 
available source of poverty data at the school level. We used these 
data as a poverty measure at the school district level because 
subsidized lunch data are available at the school, district, and state 
levels, and thus provide a consistent measure across all three levels. 
Also, subsidized lunch data are available for nearly all school 
districts, including charter schools, whereas Census poverty estimates 
are available only for the somewhat limited number of school districts 
included in Education’s database. While recognizing the limitations of 
subsidized lunch data, we believe the use of it, along with Census 
poverty estimates, strengthens our report findings. USDA’s written 
comments are printed in appendix V. 

We are sending copies of this report to the Secretaries of Education 
and Agriculture and interested congressional committees. We will also 
make copies available to others upon request. If you have any questions 
concerning this report, please contact me on (202) 512-7215. Other GAO 
contacts and staff acknowledgments are listed in appendix VI. 

Signed by: 

Marnie S. Shaul: 
Director, Education, Workforce, and Income Security Issues: 

[End of section] 

List of Congressional Addressees: 

The Honorable Tom Harkin: 
Chairman: 
The Honorable Arlen Specter: 
Ranking Minority Member: 
Subcommittee on Labor, Health and Human Services, and Education: 
Committee on Appropriations: 
United States Senate: 

The Honorable Ralph Regula: 
Chairman: 
The Honorable David Obey: 
Ranking Minority Member: 
Subcommittee on Labor, Health and Human Services, Education and Related 
Agencies: 
Committee on Appropriations: 
House of Representatives: 

The Honorable Edward M. Kennedy: 
Chairman: 
The Honorable Judd Gregg: 
Ranking Minority Member: 
Committee on Health, Education, Labor and Pensions: 
United States Senate: 

The Honorable John A. Boehner: 
Chairman: 
The Honorable George Miller: 
Ranking Minority Member: 
Committee on Education and the Workforce: 
House of Representatives: 

The Honorable Evan Bayh: 
The Honorable John Breaux: 
The Honorable Herb Kohl: 
The Honorable Mary Landrieu: 
The Honorable Joseph Lieberman: 
The Honorable Blanche Lincoln: 
United States Senate: 

[End of section] 

Appendix I: Objectives, Scope, and Methodology: 

As mandated by the Congress (Public Law 106-554 Sec. 305), we designed 
our study to provide information on (1) the extent to which Title I 
funds are allocated to states, school districts and schools with the 
greatest numbers and percentages of school-age children from low-income 
families; (2) the extent to which allocations of such funds adjust to 
shifts in numbers of children from low-income families; (3) the extent 
to which the allocation of Title I funds encourages the targeting of 
state funds to school-age children from low-income families; and (4) 
what options might improve targeting of funds, especially to states and 
school districts with higher numbers and percentages of poor children, 
to more effectively serve those children. To determine the extent to 
which Title I funds are targeted to poor children, we used two measures 
of poverty and two types of allocation data. We used the Census 
Bureau’s updated decennial poverty data for one measure. Census poverty 
data are used by the U.S. Department of Education to calculate Title I 
formula allocation amounts. For the second measure of poverty, we used 
eligibility for free or reduced-price lunches through the National 
School Lunch Program, a federal food assistance program administered by 
the U.S. Department of Agriculture for children from low-income 
families. The subsidized lunch program provides the best source of data 
on low-income students at the school-level, according to the Department 
of Education, and these data are also available for the district and 
state levels. 

We determined how states actually allocated Title I funds to each of 
their school districts by collecting 1999-2000 allocation information 
directly from state Title I officials in every state and the District 
of Columbia. In this report, we refer to these data as “actual 
allocations” or simply “allocations.” We compared the actual 
allocations with the amounts generated by the Title I formula 
calculations for the 1999-2000 school year, which we obtained from the 
Department of Education. In this report, we refer to these data as 
“formula calculations.” Formula calculations are the data typically 
used in analyses of the Title I program. The formula calculations 
accurately reflect the amount of funds allocated to each state, but 
these data do not reflect changes that states subsequently make to the 
formula-calculated amounts when allocating the funds to their school 
districts. We also interviewed state Title I directors in each of the 
50 states and the District of Columbia about their experiences and 
perceptions of the Title I program. We examined school district 
policies for allocating Title I funds to schools by surveying a 
nationally representative, stratified sample of school districts. In 
addition, we reviewed school-level allocation data from the few states 
that were able to provide it. 

We examined the responsiveness of state-level Title I allocations to 
shifts in poverty by analyzing Census data and Title I allocation data 
from the Department of Education for the period of 1980 to the present. 
We analyzed the relevant statutory provisions and reviewed our previous 
reports to identify incentives for states to target their own funds. 
Finally, we determined the consequences of various policy options by 
examining the Title I formulas and running simulations of Education’s 
formula calculation process for states and school districts for the 
2001-02 school year. We conducted our work from December 2000 to 
December 2001 in accordance with generally accepted government auditing 
standards. 

Federal Funding Formulas: 

From the Title I statute, we obtained the formulas that Education is 
required to use to calculate Title I grant amounts. We met with 
Education officials to discuss their procedures for using the formulas 
and data to calculate grant amounts. We used these procedures to 
replicate Education’s formula calculations for the 1999-2000 and 2000-
01 school years and as the basis for computer simulations of various 
changes that could be made to the formulas. 

Education’s Title I Formula Calculations: 

From the Department of Education, we obtained the grant amounts 
calculated for school districts for the 1999-2000, 2000-01, and 2001-02 
school years, which Education generated using the federal funding 
formulas included in the Title I statute. The school districts for 
which Education calculated grant amounts in the 1999-2000, 2000-01 
school years were those known to Education in the 1995-96 school year. 
The school districts for which Education calculated grant amounts in 
the 2001-02 school year were those known to Education in the 1997-98 
school year. At no time have the calculations included charter schools. 
The data set from Education also included 1995 Census data on the 
characteristics of these school districts, such as numbers of school-
age poor children, total numbers of school-age children, and total 
resident populations, which Education used in calculating grant 
amounts. The formula calculations were used to examine their 
relationship to poverty and other characteristics of school districts 
and to compare the formula calculations to the actual allocations that 
school districts received. 

Actual Allocations to School Districts: 

We collected data from state Title I program directors on the dollar 
amounts of Title I funds, if any, that they disbursed to each of their 
school districts in the 1999-2000 school year. We collected allocation 
data on basic and concentration grants to 14,682 school districts in 
all 50 states and the District of Columbia, including data on charter 
schools that are independent school districts, as they existed in the 
1999-2000 school year. Where possible, we matched the school district 
data provided by the states with the school district data provided by 
Education. Where the lists of school districts differed, we called 
state officials to verify the accuracy of their data. In most cases, 
state officials clarified that districts had been created, 
consolidated, eliminated, or had changed names since Education’s data 
were updated in 1995. For those school districts that were identified 
both by state officials and Education, we compared the actual 
allocations with Education’s formula calculations and examined the 
relationships between the actual allocations and Census poverty and 
other school district characteristics. We also used the actual 
allocation data to examine the relationship between the actual 
allocations and poverty, as measured by eligibility for free or reduced-
price lunches. 

School-Level Allocation Data: 

From state Title I directors, we requested data on the dollar amount of 
Title I funds that each of their states’ schools received in the 1999-
2000 school year, if these data were available in an electronic format. 
From state food services officials, we also requested for each school, 
electronic data on enrollment and the number of students receiving free 
or reduced-price lunches. Only three states (California, Georgia, and 
Mississippi) could provide us with school-level allocations, 
enrollment, and school lunch data in an electronic format for each of 
their schools. 

For each school in these three states, we matched the allocation data 
to the enrollment and school lunch data and calculated both the 
percentages of children eligible for free or reduced-price lunch and 
the amount of funds received per poor child. Because we obtained data 
on every school within these three states, there is no estimation or 
sampling error associated with our results. However, our findings based 
on these data are not generalizable beyond the state or school year for 
which the data were collected. 

Free and Reduced-Price Lunch Data: 

To estimate the numbers and percentages of children in poverty in every 
school district, we obtained data from state school food service 
officials on both the numbers of children receiving free or reduced-
price lunches through the National School Lunch Program (NLSP) and the 
total number of students enrolled in each school district in the 1999-
2000 school year. We also obtained these data at the school level, 
where available. Children from families with incomes at or below 130 
percent of the poverty level are eligible for free meals through NSLP; 
those with incomes between 130 and 185 percent of the poverty level are 
eligible for reduced-price meals. 

We requested NSLP participation data for every school district in every 
state, including charter schools that are independent school districts, 
where available. These data were combined with school district-level 
data on Title I allocations in order to calculate the amount of Title I 
allocations per poor child received in each district. 

We chose participation in the NSLP as a measure of school district 
poverty because it is the measure used most commonly by school 
districts to determine allocations to schools and is the best source of 
poverty data that is available at the state, district, and school 
levels. Because participation in the NSLP is voluntary, there is some 
concern that participation rates may reflect, in part, the effort 
schools make to encourage participation, and may not consistently 
reflect actual program eligibility rates across schools and school 
districts. There is also concern that high school students are less 
likely to participate in the program than younger students due to the 
associated stigma. Nonetheless, a National Research Council panel 
concluded that NSLP participation is an indicator of low family income 
and that the quality of NSLP data are neither appreciably better nor 
worse than Census data for measuring poverty, especially for areas as 
small as school districts. 

Waivers of Title I Regulations: 

To obtain information on the number and types of waivers granted to 
districts under the Elementary and Secondary Education Act, we reviewed 
Education’s annual reports to the Congress for 1998-2001 and met with 
program officials. To obtain information on the number and types of 
waivers granted by “Ed-Flex” states under the 1994 and 1996 Ed-Flex 
Demonstration Project and the Ed-Flex Partnership Act of 1999, we met 
with program officials and reviewed states’ Ed-Flex applications on 
file with Education. We used this information to determine the most 
common types of waivers that were granted overall. 

Interviews With State Title I Directors: 

To obtain information on both states’ roles in the Title I allocation 
process and the opinions of state Title I directors, we conducted 
telephone interviews with the directors of the Title I program in every 
state and the District of Columbia between December 2000 and May 2001, 
using a semi-structured interview protocol. We asked the directors to 
explain exactly how they generate dollar allocations to school 
districts once they receive the information from the Department of 
Education, including how they apportion funds to small size districts 
and districts whose boundaries have changed, how charter schools are 
handled in the allocation process, what data they use, and how recent 
those data are. Finally, we asked whether their states had compensatory 
education programs that target funding to high-poverty schools and 
districts and, if so, how allocations for that program were related to 
the allocation of Title I dollars. 

Survey of School Districts: 

We surveyed a stratified nationally representative sample of school 
district administrators drawn from the approximately 13,000 school 
districts nationwide for which Education had calculated an initial 
Title I allocation amount for the 1999-2000 school year. In addition to 
providing information on their school districts’ schools and 
communities, survey respondents provided information on how they 
measure poverty in their schools, their priorities and rationales in 
distributing funds, and their use of funds for district-level 
activities. 

The sample was stratified into four categories according to the number 
of school-age children living in the school district boundaries, as 
follows: 2-500 children; 501-2,500 children; 2501-50,000 children; over 
50,000 children. A random sample was drawn from each of the first three 
strata; all of the 96 school districts with greater than 50,000 
children were included in the survey. Table 14 provides information on 
the total numbers of students and districts, the number of districts 
sampled, and the response rate for each of the strata. This sample 
design allows us to generalize our results to all school districts of 
similar sizes, including the very smallest school districts. The survey 
was conducted between July and October 2001 and reflects school 
district decisions in the 2000-01 school year. 

Table 14: Number of Students, Districts, Districts Sampled, and 
Response Rates by Survey Strata: 

Strata (by number of students in district): 2-500 students: 
Number of students nationwide (percentage of all students): 876,634 
(0.2%); 
Number of districts nationwide[A] (percentage of all districts): 3,385 
(26%); 
Number of districts sampled: 173; 
Response rate (percentage): 92. 

Strata (by number of students in district): 501-2,500 students: 
Number of students nationwide (percentage of all students): 7,118,365 
(15%); 
Number of districts nationwide[A] (percentage of all districts): 5,660 
(44%); 
Number of districts sampled: 187; 
Response rate (percentage): 94. 

Strata (by number of students in district): 2,501-50,000 students: 
Number of students nationwide (percentage of all students): 28,576,493 
(61%); 
Number of districts nationwide[A] (percentage of all districts): 3,741 
(29%); 
Number of districts sampled: 397; 
Response rate (percentage): 91. 

Strata (by number of students in district): More than 50,000 students: 
Number of students nationwide (percentage of all students): 10,295,794 
(22%); 
Number of districts nationwide[A] (percentage of all districts): 96 
(0.7%); 
Number of districts sampled: 93; 
Response rate (percentage): 89. 

Total: 
Number of students nationwide (percentage of all students): 46,867,286 
(100%); 
Number of districts nationwide[A] (percentage of all districts): 12,882 
(100%); 
Number of districts sampled: 850; 
Response rate (percentage): 92. 

Note: An estimated 20 percent of school districts consist of only one 
school, so survey questions regarding how funds were distributed among 
schools were not relevant. 

[A] Districts sampled included only those identified by Education. 

Source: GAO Analysis. 

[End of table] 

Because our estimates are based on samples, they are subject to 
sampling error. Table 15 shows each of our estimates and indicates the 
extent of each estimate’s sampling error by showing the 95-percent 
confidence interval around that estimate. There is a 95-percent chance 
that the actual total falls within the interval. 

Table 15: Sampling Errors: 

Percentage of School districts[A] prioritizing primary schools: 
Sampling error[B] (percentage points): 4. 

Percentage of School districts prioritizing poverty rankings: 
Sampling error[B] (percentage points): 3. 

Percentage of School districts preserving funds for districtwide 
activities: 
Sampling error[B] (percentage points): 5. 

Percentage of Urban school districts prioritizing primary schools: 
Sampling error[B] (percentage points): 14. 

Percentage of Urban school districts prioritizing poverty rankings: 
Sampling error[B] (percentage points): 13. 

Percentage of Rural school districts prioritizing primary schools: 
Sampling error[B] (percentage points): 6. 

Percentage of Rural school districts prioritizing poverty rankings 
Sampling error[B] (percentage points): 4. 

[A] School district officials self-identified their districts as urban, 
rural, suburban, or mixed. 

[B] At the 95-percent confidence interval. 

Source: GAO Analysis. 

[End of table] 

Interviews With Federal Officials and Education Experts: 

During our work, we consulted with representatives from the following 
agencies and organizations who have knowledge of the Title I program 
and related issues: U.S. Department of Education, U.S. Department of 
Agriculture, National Research Council, U.S. Census Bureau’s Small-Area 
Income and Poverty Estimates Panel, Congressional Research Service, 
Council of Chief State School Officers, Center on Education Policy, 
American Association of School Administrators, Council of the Great 
City Schools, and the National Association of State Title I Directors. 

[End of section] 

Appendix II: Supporting Data: 

Variation in Funding Per Child Under Alternative Hold-Harmless Rules by 
Poverty Groups: 

Local school districts were put into one of five groups with each group 
containing an equal number of poor children. The groups ranged from the 
lowest percentages of poverty to the highest percentages of poverty. 
Each group represents approximately 20 percent of all formula eligible 
children.[Footnote 44] For this analysis, we expressed the number of 
formula-eligible children as a percentage of the number of children 
ages 5-17. Table 16 shows the average funding per child allotted by 
formula, formula allotments under each of the simulations, and percent 
differences in funding per child compared to 2001 formula allotments. 

Table 16: District Funding Per Child by Poverty Group: 

Poverty group: 2001 formula allocation: Dollars per child; 
Lowest (no more than 15% poverty): $758; 
Low (15-21% poverty): $821; 
Medium (21-29% poverty): $826; 
High (29-40% poverty): $865; 
Highest (40-100% poverty): $864; 
Ratio (highest to lowest): 114%. 

Poverty group: 100-percent hold-harmless: Dollars per child; 
Lowest (no more than 15% poverty): $751; 
Low (15-21% poverty): $816; 
Medium (21-29% poverty): $824; 
High (29-40% poverty): $869; 
Highest (40-100% poverty): $867; 
Ratio (highest to lowest): 115%. 

Poverty group: 100-percent hold-harmless: Percentage difference; 
Lowest (no more than 15% poverty): -0.8%; 
Low (15-21% poverty): -0.6%; 
Medium (21-29% poverty): -0.2%; 
High (29-40% poverty): 0.5%; 
Highest (40-100% poverty): 0.3%. 

Poverty group: Authorizing hold-harmless: Dollars per child; 
Lowest (no more than 15% poverty): $734; 
Low (15-21% poverty): $826; 
Medium (21-29% poverty): $823; 
High (29-40% poverty): $867; 
Highest (40-100% poverty): $878; 
Ratio (highest to lowest): 120%. 

Poverty group: Authorizing hold-harmless: Percentage difference; 
Lowest (no more than 15% poverty): -3.1%; 
Low (15-21% poverty): 0.6%; 
Medium (21-29% poverty): -0.4%; 
High (29-40% poverty): 0.3%; 
Highest (40-100% poverty): 1.6%. 

Poverty group: No hold-harmless: Dollars per child; 
Lowest (no more than 15% poverty): $739; 
Low (15-21% poverty): $829; 
Medium (21-29% poverty): $820; 
High (29-40% poverty): $853; 
Highest (40-100% poverty): $885; 
Ratio (highest to lowest): 120%. 

Poverty group: No hold-harmless: Percentage difference; 
Lowest (no more than 15% poverty): -2.4%; 
Low (15-21% poverty): 0.9%; 
Medium (21-29% poverty): -0.7%; 
High (29-40% poverty): -1.3%; 
Highest (40-100% poverty): 2.5%. 

Note: Even though these groups represent equal percentages of children, 
they represent unequal ranges in terms of percentage of poverty. 

Source: GAO Analysis. 

[End of table] 

Table 17 shows the coefficients of variation for figure 5. 

Table 17: Variation in Funding Per Child Within Poverty Groups: 

Poverty group: 2001 formula coefficient of variation (CoV); 
Lowest (no more than 15% poverty): 30%; 
Low (15-21% poverty): 21%; 
Medium (21-29% poverty): 19%; 
High (29-40% poverty): 17%; 
Highest (40-100% poverty): 19%. 

Poverty group: 100-percent hold-harmless: CoV; 
Lowest (no more than 15% poverty): 32%; 
Low (15-21% poverty): 21%; 
Medium (21-29% poverty): 20%; 
High (29-40% poverty): 18%; 
Highest (40-100% poverty): 19%. 

Poverty group: 100-percent hold-harmless: Percentage difference[A]; 
Lowest (no more than 15% poverty): 3.8%; 
Low (15-21% poverty): 4.1%; 
Medium (21-29% poverty): 3.3%; 
High (29-40% poverty): 3.3%; 
Highest (40-100% poverty): 3.2%. 

Poverty group: Authorizing hold-harmless: CoV; 
Lowest (no more than 15% poverty): 20%; 
Low (15-21% poverty): 14%; 
Medium (21-29% poverty): 15%; 
High (29-40% poverty): 15%; 
Highest (40-100% poverty): 16%. 

Poverty group: Authorizing hold-harmless: Percentage difference[A]; 
Lowest (no more than 15% poverty): -33.3%; 
Low (15-21% poverty): -30.0%; 
Medium (21-29% poverty): -21.2%; 
High (29-40% poverty): -13.6%; 
Highest (40-100% poverty): -15.6%. 

Poverty group: No hold-harmless: CoV; 
Lowest (no more than 15% poverty): 16%; 
Low (15-21% poverty): 13%; 
Medium (21-29% poverty): 14%; 
High (29-40% poverty): 14%; 
Highest (40-100% poverty): 16%. 

Poverty group: No hold-harmless: Percentage difference[A]; 
Lowest (no more than 15% poverty): -48.9%; 
Low (15-21% poverty): -34.3%; 
Medium (21-29% poverty): -28.9%; 
High (29-40% poverty): -17.1%; 
Highest (40-100% poverty): -15.8%. 

[A] Percentage differences calculated based on unrounded numbers. 

Source: GAO Analysis. 

[End of table] 

Table 18 shows the data used to construct figure 6. 

Table 18: Percentage Change in Funding With No Hold-Harmless, Analysis 
by Changes in State Poverty 2001-2002 (States Ranked by Percentage 
Growth in the Number of Poor Children): 

State: Alaska; 
Percentage change in formula-eligible children: 39.6; 
Percentage change in Title I funding: 0.9. 

State: Utah; 
Percentage change in formula-eligible children: 35.2; 
Percentage change in Title I funding: -6.8; 

State: Nevada; 
Percentage change in formula-eligible children: 24.6; 
Percentage change in Title I funding: 6.9. 

State: New Hampshire; 
Percentage change in formula-eligible children: 22.1; 
Percentage change in Title I funding: -14.3. 

State: Massachusetts; 
Percentage change in formula-eligible children: 20.1; 
Percentage change in Title I funding: 1.8. 

State: Hawaii; 
Percentage change in formula-eligible children: 17.8; 
Percentage change in Title I funding: 7.0. 

State: Maryland; 
Percentage change in formula-eligible children: 16.4; 
Percentage change in Title I funding: 4.6. 

State: New Jersey; 
Percentage change in formula-eligible children: 14.8; 
Percentage change in Title I funding: 2.4. 

State: Colorado; 
Percentage change in formula-eligible children: 13.5; 
Percentage change in Title I funding: 2.9. 

State: Idaho; 
Percentage change in formula-eligible children: 12.5; 
Percentage change in Title I funding: -2.3. 

State: Minnesota; 
Percentage change in formula-eligible children: 11.9; 
Percentage change in Title I funding: 1.4. 

State: Washington; 
Percentage change in formula-eligible children: 11.0; 
Percentage change in Title I funding: 1.9. 

State: Connecticut; 
Percentage change in formula-eligible children: 10.8; 
Percentage change in Title I funding: 5.6. 

State: California; 
Percentage change in formula-eligible children: 8.7; 
Percentage change in Title I funding: 4.3. 

State: Virginia; 
Percentage change in formula-eligible children: 8.7; 
Percentage change in Title I funding: 6.2. 

State: Oregon; 
Percentage change in formula-eligible children: 8.6; 
Percentage change in Title I funding: 3.1. 

State: Delaware; 
Percentage change in formula-eligible children: 8.4; 
Percentage change in Title I funding: -16.5. 

State: Arizona; 
Percentage change in formula-eligible children: 6.4; 
Percentage change in Title I funding: 4.1. 

State: Indiana; 
Percentage change in formula-eligible children: 5.9; 
Percentage change in Title I funding: 3.3. 

State: Kansas; 
Percentage change in formula-eligible children: 4.9; 
Percentage change in Title I funding: 3.2. 

State: Rhode Island; 
Percentage change in formula-eligible children: 4.4; 
Percentage change in Title I funding: 6.2. 
 
State: Iowa; 
Percentage change in formula-eligible children: 3.3; 
Percentage change in Title I funding: -9.6. 

State: Georgia; 
Percentage change in formula-eligible children: 1.4; 
Percentage change in Title I funding: 6.4. 

State: Wyoming; 
Percentage change in formula-eligible children: 1.3; 
Percentage change in Title I funding: -9.1. 

State: North Dakota; 
Percentage change in formula-eligible children: 1.0; 
Percentage change in Title I funding: -10.1. 

State: Nebraska; 
Percentage change in formula-eligible children: 0.2; 
Percentage change in Title I funding: -13.2. 

State: Illinois; 
Percentage change in formula-eligible children: -0.1; 
Percentage change in Title I funding: 2.2. 

State: New York; 
Percentage change in formula-eligible children: -0.3; 
Percentage change in Title I funding: 3.8. 

State: North Carolina; 
Percentage change in formula-eligible children: -0.4; 
Percentage change in Title I funding: 5.2. 

State: Montana; 
Percentage change in formula-eligible children: -0.9; 
Percentage change in Title I funding: -0.8. 

State: Wisconsin; 
Percentage change in formula-eligible children: -1.0; 
Percentage change in Title I funding: -3.2. 

State: Florida; 
Percentage change in formula-eligible children: -2.1; 
Percentage change in Title I funding: 5.6. 

State: South Carolina; 
Percentage change in formula-eligible children: -2.3; 
Percentage change in Title I funding: 3.7. 

State: Michigan; 
Percentage change in formula-eligible children: -3.1; 
Percentage change in Title I funding: -3.5. 

State: Pennsylvania; 
Percentage change in formula-eligible children: -5.1; 
Percentage change in Title I funding: -4.8. 

State: Oklahoma; 
Percentage change in formula-eligible children: -6.6; 
Percentage change in Title I funding: -1.2. 

State: Alabama; 
Percentage change in formula-eligible children: -6.9; 
Percentage change in Title I funding: -5.2. 

State: Ohio; 
Percentage change in formula-eligible children: -7.1; 
Percentage change in Title I funding: -10.7. 

State: South Dakota; 
Percentage change in formula-eligible children: -7.2; 
Percentage change in Title I funding: -4.7. 

State: Arkansas; 
Percentage change in formula-eligible children: -7.3; 
Percentage change in Title I funding: -4.5. 

State: Missouri; 
Percentage change in formula-eligible children: -7.7; 
Percentage change in Title I funding: -4.0. 

State: Kentucky; 
Percentage change in formula-eligible children: -7.8; 
Percentage change in Title I funding: -3.8. 

State: Texas; 
Percentage change in formula-eligible children: -8.1; 
Percentage change in Title I funding: -0.4. 

State: Maine; 
Percentage change in formula-eligible children: -8.7; 
Percentage change in Title I funding: -9.2. 

State: District of Columbia; 
Percentage change in formula-eligible children: -8.8; 
Percentage change in Title I funding: 6.3. 

State: Tennessee; 
Percentage change in formula-eligible children: -11.2; 
Percentage change in Title I funding: -13.5. 

State: Vermont; 
Percentage change in formula-eligible children: -11.3; 
Percentage change in Title I funding: -8.4. 

State: New Mexico; 
Percentage change in formula-eligible children: -11.7; 
Percentage change in Title I funding: 0.7. 

State: Louisiana; 
Percentage change in formula-eligible children: -16.0; 
Percentage change in Title I funding: -13.8. 

State: West Virginia; 
Percentage change in formula-eligible children: -17.1; 
Percentage change in Title I funding: -11.9. 

State: Mississippi; 
Percentage change in formula-eligible children: -19.9; 
Percentage change in Title I funding: -28.9. 

Source: GAO Analysis. 

[End of table] 

Replacing State Per-We analyzed three formula scenarios that replaced 
the state per-pupil spending factor with an alternative cost factor 
developed by the Pupil Expenditure Department of Education: a state-
level cost factor, a district-level cost Factor With a Cost of factor, 
and a district-level cost factor combined with the hold-harmless rules 
described in the authorizing statute. Table 19 reports the average 
Education Factor funding per child in each poverty group and table 20 
reports the coefficients of variation in funding per child within each 
group that were reported in figure 7 of the report. 

Table 19: Funding Per Child by Poverty Group Under Formula Options That 
Replace State Per-Pupil Spending With a Cost of Education Factor: 

Poverty group: 2001 formula allocation: Dollars per child; 
Lowest (no more than 15% poverty): $758; 
Low (15-21% poverty): $821; 
Medium (21-29% poverty): $826; 
High (29-40% poverty): $865; 
Highest (40-100% poverty): $864; 
Ratio (highest to lowest): 114%. 

Poverty group: State cost factor: Dollars per child; 
Lowest (no more than 15% poverty): $737; 
Low (15-21% poverty): $815; 
Medium (21-29% poverty): $825; 
High (29-40% poverty): $866; 
Highest (40-100% poverty): $892; 
Ratio (highest to lowest): 121%. 

Poverty group: State cost factor: Percentage difference; 
Lowest (no more than 15% poverty): -2.7%; 
Low (15-21% poverty): -0.8%; 
Medium (21-29% poverty): 0.0%; 
High (29-40% poverty): 0.1%; 
Highest (40-100% poverty): 3.3%. 

Poverty group: District cost factor: Dollars per child; 
Lowest (no more than 15% poverty): $746; 
Low (15-21% poverty): $815; 
Medium (21-29% poverty): $821; 
High (29-40% poverty): $866; 
Highest (40-100% poverty): $887; 
Ratio (highest to lowest): 119%. 

Poverty group: District cost factor: Percentage difference; 
Lowest (no more than 15% poverty): -1.6%; 
Low (15-21% poverty): -0.8%; 
Medium (21-29% poverty): -0.5%; 
High (29-40% poverty): 0.1%; 
Highest (40-100% poverty): 2.7%. 

Poverty group: District cost factor & authorizing hold-harmless: 
Dollars per child; 
Lowest (no more than 15% poverty): $716; 
Low (15-21% poverty): $818; 
Medium (21-29% poverty): $818; 
High (29-40% poverty): $873; 
Highest (40-100% poverty): $905; 
Ratio (highest to lowest): 126%. 

Poverty group: District cost factor & authorizing hold-harmless: 
Percentage difference; 
Lowest (no more than 15% poverty): -5.4%; 
Low (15-21% poverty): -0.5%; 
Medium (21-29% poverty): -1.0%; 
High (29-40% poverty): 0.9%; 
Highest (40-100% poverty): 4.7%. 

Source: GAO Analysis. 

[End of table] 

Table 20: Variation in Funding Per Child Within Poverty Groups Under 
Formula Options That Replace State Per-Pupil Spending With a Cost of 
Education Factor: 

Poverty group: 2001 formula coefficient of variation (CoV); 
Lowest (no more than 15% poverty): 30%; 
Low (15-21% poverty): 21%; 
Medium (21-29% poverty): 19%; 
High (29-40% poverty): 17%; 
Highest (40-100% poverty): 19%. 

Poverty group: State cost factor: CoV; 
Lowest (no more than 15% poverty): 30%; 
Low (15-21% poverty): 19%; 
Medium (21-29% poverty): 17%; 
High (29-40% poverty): 15%; 
Highest (40-100% poverty): 13%. 

Poverty group: State cost factor: Percentage difference; 
Lowest (no more than 15% poverty): -0.5%; 
Low (15-21% poverty): -5.9%; 
Medium (21-29% poverty): -10.1%; 
High (29-40% poverty): -13.3%; 
Highest (40-100% poverty): -29.7%. 

Poverty group: District cost factor: CoV; 
Lowest (no more than 15% poverty): 30%; 
Low (15-21% poverty): 20%; 
Medium (21-29% poverty): 18%; 
High (29-40% poverty): 16%; 
Highest (40-100% poverty): 14%. 

Poverty group: District cost factor: Percentage difference; 
Lowest (no more than 15% poverty): -1.3%; 
Low (15-21% poverty): -3.7%; 
Medium (21-29% poverty): -4.3%; 
High (29-40% poverty): -9.2%; 
Highest (40-100% poverty): -25.0%. 

Poverty group: District cost factor & authorizing hold-harmless: CoV; 
Lowest (no more than 15% poverty): 19%; 
Low (15-21% poverty): 12%; 
Medium (21-29% poverty): 13%; 
High (29-40% poverty): 13%; 
Highest (40-100% poverty): 10%. 

Poverty group: District cost factor & authorizing hold-harmless: 
Percentage difference; 
Lowest (no more than 15% poverty): -36.1%; 
Low (15-21% poverty): -40.1%; 
Medium (21-29% poverty): -30.4%; 
High (29-40% poverty): -26.3%; 
Highest (40-100% poverty): -44.6%. 

Source: GAO Analysis. 

[End of table] 

[End of section] 

Appendix III: Selected Provisions of the “No Child Left Behind Act of 
2001” and Related Appropriations: 

On January 8, 2002, President Bush signed into law the “No Child Left 
Behind Act of 2001,” reauthorizing Title I and other Elementary and 
Secondary Education Act programs, with some significant changes. 
[Footnote 45] Two days later, he signed the related appropriations law. 
[Footnote 46] The changes to Title I relevant to this report are 
outlined below: 

Funding for Targeted Grant Formula: For the first time, the Congress 
appropriated funds for targeted grants. The 2001 Act requires that the 
amounts allocated through basic and concentration grants are to be the 
same as they were in fiscal year 2001 and that any additional funds 
remaining (i.e., any new funds) are to be allocated through the 
targeted grant formula. As under prior law, a tiered weighting system 
would provide proportionately greater funding per poor child to 
districts with higher numbers and percentages of poor children. The new 
law changed the cut-points between the tiers slightly, based on updated 
Census poverty estimates, so that each tier would continue to contain 
roughly equal numbers of poor children. 

Funding for Revised Finance Incentive Grant Formula: Not only was the 
finance incentive grant funded for the first time, but the grant 
formula and other provisions also were significantly revised. In prior 
law, the incentive grant formula was designed to provide additional 
funds to states that demonstrated high state education spending 
relative to their tax base and states that had less disparity in 
funding among districts. The new law maintains these provisions and 
adds several more that give proportionately more funds to states and 
districts with higher numbers and percentages of poor children, as 
follows: 

* Allocations will be based on each state’s number of poor children, 
rather than its total school-age population. 

* Districts are required to have at least 10 poor children, making up 
at least 5 percent of enrollment to qualify for finance incentive 
funds, whereas there had been no such enrollment requirement in prior 
law. 

* Allocations will be made to school districts on the basis of a tiered 
weighting system, like that in the targeted grant formula. The 
incentive grant weighting system provides proportionally more funds not 
only to districts with greater numbers and percentages of poor children 
but also to districts in states with less funding disparity among 
districts. 

* Districts are newly required to allocate finance incentive funds to 
schools in the same way that they allocate the other Title I funds 
(e.g., in rank order of poverty) and to use finance incentive funds 
only for Title I purposes. 

In addition, the new formula includes a per-pupil expenditure factor, 
like that for the other grants, but more narrowly limited to a minimum 
of 34 percent and a maximum of 46 percent of the national average per 
pupil expenditure, rather than the 32 percent minimum and 48 percent 
maximum in the other grant formulas. 

Increased Overall Funding for Fiscal Year 2002: The education 
appropriations legislation includes a combined increase of nearly $1.8 
billion in funding for Title I basic, concentration, targeted, and 
finance incentive grants. (See table 21) 

Table 21: Change in Title I Appropriations Between Fiscal Years 2001 
and 2002 (Thousands of dollars): 

Basic grants: 
FY 2001: $7,237,721; 
FY 2002: $7,172,971; 
Percent increase FY 2001-2002: -0.89. 

Concentration grants:
FY 2001: $1,364,000; 
FY 2002: $1,365,031; 
Percent increase FY 2001-2002: 0.08. 

Targeted grants: 
FY 2001: 0; 
FY 2002: $1,018,499; 
Percent increase FY 2001-2002: N/A. 

Finance incentive grants: 
FY 2001: 0; 
FY 2002: $793,499; 
Percent increase FY 2001-2002: N/A. 

Total: 
FY 2001: $8,601,721; 
FY 2002: $10,350,000; 
Percent increase FY 2001-2002: 20.32. 

Source: Public Law 106-554 (Consolidated Appropriations Act, 2001), 
Public Law 107-20 (Supplemental Appropriations Act, 2001), and Public 
Law 107-116 (Appropriations Act, 2002). 

[End of table] 

Hold-Harmless: The prior authorizing legislation included a hold-
harmless provision only for basic grants (districts were guaranteed 85, 
90, or 95 percent of the previous year’s funding, depending on 
percentage of poor children in the district). However, as described in 
the body of the report, appropriations language in recent years created 
more restrictive hold-harmless provisions, including (1) a 100-percent 
hold-harmless provision for basic grants and (2) a hold-harmless 
provision for concentration grants that allowed even districts no 
longer meeting the concentration grant eligibility criteria to continue 
receiving concentration grants. In contrast, the fiscal year 2002 
appropriations language does not include any provisions that override 
the authorized hold-harmless provisions. However, under the new 
authorizing legislation, hold-harmless provisions will apply not only 
to basic grants but also to concentration grants and targeted grants. 
In addition, under the new authorizing legislation, districts that 
become ineligible for concentration grants will continue to receive 
concentration grant allocations for up to 4 consecutive years. 

As a result of these changes, operative hold-harmless provisions for 
basic grants are somewhat less restrictive than under prior law and, 
therefore, will allow basic grant allocations to be more reflective of 
the number of poor children in a school district. Likewise, the newly 
authorized hold-harmless provisions for concentration grants are 
somewhat less restrictive than the hold-harmless provisions previously 
included in appropriations law. In the end, allocations under these 
grants will be more reflective of numbers of poor children than in the 
past, but not as reflective of them as would have been the case under 
the prior authorizing legislation alone. 

Small State Minimum: The minimum level of funding guaranteed to each 
state is increased over prior law. Previously, each state was 
guaranteed the smaller of 0.25 percent of total appropriations for that 
year, or the average of that amount and the state’s number of eligible 
students multiplied by 150 percent of the national average per-pupil 
payment. The new law uses essentially the same calculation but instead 
of 0.25 percent of the total appropriations, the new calculation will 
use 0.25 percent of the amount appropriated in 2001, plus 0.35 percent 
of any subsequent increases in appropriations over the 2001 level. 

Other Funding Details: 

* States are required to reserve 2 percent of their Title I funds for 
school improvement, increasing to 4 percent in fiscal year 2004. 
Previously, states were permitted, but not required, to reserve up to 
0.5 percent of their funds for school improvement. States must pass at 
least 95 percent of these funds directly to school districts. 

* Districts must continue to reserve at least 1 percent of their Title 
I allocations for parental involvement activities, as was required 
under prior law, but the new law also requires that they pass 95 
percent of these reserved funds to Title I schools. 

* Districts may use Title I funds for schoolwide programs, rather than 
targeting funds to specific students, in schools where at least 40 
percent of the children in the school or school attendance area are 
from low-income families. Previously, schoolwide programs were allowed 
only in schools in which at least 50 percent of the children in the 
school or school attendance area were from low-income families. 

[End of section] 

Appendix IV: Comments From the Department of Education: 

United States Department Of Education: 
The Deputy Secretary: 
400 Maryland Ave. S.W. 
Washington, D.C. 20202-0500: 
"Our mission is to ensure equal access to education and to promote 
educational excellence throughout the Nation." 

January 18, 2002: 

Marnie S. Shaul: 
Director: 
Education, Workforce, and Income Security Issues: 
U.S. General Accounting Office: 
Washington, DC 20548: 

Dear Ms. Shaul: 

We have reviewed GAO's draft report entitled Title I Funding: Poor 
Children Benefit Though Funding Per Poor Child Differs and we 
appreciate the thorough and useful analysis carried out by GAO staff. 

We agree with GAO's finding that Title I funds are generally targeted 
to higher-poverty districts and schools but that this targeting could 
be improved. In particular, as the GAO report points out, hold-harmless 
provisions have limited the extent to which Title I funds follow poor 
children when geographic shifts in poverty occur. Although the Census
Bureau now provides updated poverty data every two years (and the new 
reauthorization calls for annual updates), enabling the Department to 
use more current data for allocating Title I funds, these new data have 
had little effect on Title I allocations due to the restrictions 
imposed by 100 percent hold-harmless provisions included in every 
appropriations act since fiscal year 1997 (with the exception of the 
most recent appropriations for fiscal year 2002). As a result, States 
that have experienced growth in their numbers of poor children have 
suffered declines in their relative shares of Title I funding per poor 
child. 

The GAO report points out the wide disparities across States in Title I 
funding per poor child, ranging from a high of $1,150 in Vermont to a 
low of $520 in New Mexico. Indeed, the Nation's poorest States receive 
some of the lowest allocations per poor child. GAO notes that these 
disparities are partly due to the use of State per-pupil expenditure 
factor (intended to adjust for interstate differences in the cost of 
education), as well as small-State minimum and hold-harmless 
provisions. GAO suggests that an alternative cost indicator would 
direct more funding to districts with higher percentages of poor 
children and might more accurately reflect educational costs. Although 
the Department has not fully examined GAO's specific suggestions for 
increasing the targeting of Title I funds, we strongly support the 
general goal expressed in the No Child Left Behind Act of "targeting 
resources sufficiently to make a difference to local educational 
agencies and schools where the needs are greatest." 

The report should be updated to reflect the fiscal 2002 appropriations 
act which funded the Targeted Grants and Incentive Grants formulas for 
the first time and, unlike previous years' acts, did not include hold-
harmless language. It should also reflect the passage of the No Child 
Left Behind Act, which revised the Incentive Grants formula to allocate 
funds on the basis of poverty, consistent with the other Title I 
formulas. 

Thank you for the opportunity to comment on this draft report. 

Sincerely, 

Signed by: 

William D. Hansen: 

[End of section] 

Appendix V: Comments From the U.S. Department of Agriculture: 

USDA: 
United States Department of Agriculture: 
Food and Nutrition Service: 
3101 Park Center Drive: 
Alexandria, VA 22302-1500: 

January 22, 2002: 

To: Marnie Shaul: 
Director: 
Education, Workforce, and Income Security Issues: 
General Accounting Office: 

From: [Signed by] George A. Braley: 
Acting Administrator: 

Subject: Comments on GAO Report: "Title I Funding: Poor Children 
Benefit Though Funding Per Poor Child Differs." 

An objective of this report is to assess the extent to which Title I 
funds are allocated to states, school districts, and schools with the 
greatest numbers and percentages of school-age poor children. In its 
analysis, GAO uses data on approvals for free and reduced price meals 
in the National School Lunch Program (NSLP) as a measure of poverty at 
both the school and district level. The Food and Nutrition Service 
(FNS) recognizes that NSLP free and reduced price data is often used in 
this manner for both analytical and operational purposes. However, FNS 
has become increasingly aware of the limitations of this data as a 
measure of low income status. Information from a variety of sources 
indicates that many students who are above the NSLP income eligibility 
limits (130 and 185 percent of poverty) are approved for free or 
reduced price meals. We believe that this has important implications 
for the targeting of Title I funds which GAO should consider in this 
report. 

School Level Allocations: Free and reduced price data is overwhelmingly 
the predominant source used by districts to allocate Title I funds 
among schools. GAO's current draft states that "the subsidized school 
lunch program provides the best available source of data on low-income 
students at the school level" and does not comment further on its 
accuracy for school-level allocations. We believe, however, that error 
in NSLP data is likely contributing to inaccuracies in Title I funds 
allocation, particularly because of the "threshold" approach reflected 
in the district allocation process. Schools above a given percentage of 
enrollment approved for free and/or reduced price meals receive funding 
while those below this percentage do not. If some schools only meet the 
threshold because ineligible students are included in the free and 
reduced price counts, then funds are being diverted from needier 
schools whose true eligibility levels meet the threshold. 

District Level Allocations: NSLP free and reduced price data are not 
important determinants of district-level Title I allocations. However, 
the draft report does use these data as a measure of district level 
poverty, and notes in several places that such data are closely linked 
to district-level poverty levels. FNS does not believe that this 
relationship is as strong or consistent as a reader of this draft might 
conclude. For example, FNS will soon be releasing data on free and 
reduced approvals which suggests significant variation in its accuracy 
(and thus its accuracy as a proxy for poverty) among school districts. 
Further, the NAS study cited in this report as evidence of this 
relationship was based on a very small sample of districts, and the NAS 
itself noted that further assessment is needed. 

To summarize, we believe that in order to present a full assessment of 
the accuracy of Title I targeting to low-income schools and districts, 
the report should address these issues. We would be happy to discuss 
these comments with you further. 

[End of section] 

Appendix VI: GAO Contacts and Staff Acknowledgments: 

GAO Contacts: 

Jeff Appel, (202) 512-9915: 
Heather McCallum, (202) 512-2890: 

Staff Acknowledgments: 

In addition to those named above, the following people made significant 
contributions to this report: Natalie Britton, Karen Brown, Patrick 
DiBattista, Robert Dinkelmeyer, Jerry Fastrup, Sarah Glavin, Sonya 
Harmeyer, Peter Minarik, and Michael Williams. Jon Barker, Richard 
Burkard, and Robert Parker also provided key technical assistance. 

[End of section] 

Footnotes: 

[1] Throughout this report, we refer to Title I, Part A of the 
Elementary and Secondary Education Act as “Title I.” Other Parts of 
Title I (Parts B, C, and D) are targeted at specific populations and 
are commonly referred to by their program names (Even Start, Migrant 
Education, and grants for Neglected and Delinquent Children). 

[2] The districts appearing in Education’s database for the 1999-2000 
school year were the districts as they were configured in 1995. 

[3] In March 2001, we briefed officials on the Authorizing and 
Appropriations Committees on the preliminary results of this work. 

[4] In addition to poverty, standards used to classify children as 
Title I-eligible include participation in the Temporary Assistance for 
Needy Families (TANF) program and being a child in a foster home or in 
a locally operated institution for neglected and delinquent children. 
Collectively, these children are called “formula-eligible children.” 
Unless otherwise noted, in this report, “poor children” is used to 
refer to children who are poor as measured by Census data or, when 
indicated, children eligible for free or reduced-price lunch. Ninety-
six percent of all children classified as formula-eligible would also 
be classified as eligible for Title I using Census poverty estimates. 

[5] Although the districts receiving basic grants and concentration 
grants contain about 99 and 85 percent of poor children, respectively, 
only about 27 percent of all public school students are affected by 
Title I services. School districts distribute their allocations to a 
limited number of their schools, which provide services to a limited 
number of their students. 

[6] Legislation signed by President Bush in January 2002 funds these 
grants beginning in school year 2002-03. 

[7] Legislation signed by President Bush in January 2002 funds these 
grants beginning in school year 2002-03. 

[8] In other words, expenditures would be limited to a minimum of 32 
percent and a maximum of 48 percent of the national average 
expenditure. 

[9] For concentration grants, that last term in the average is the 
greater of the state’s number of eligible children multiplied by 150 
percent of the national average per-pupil payment, or the amount 
$340,000. 

[10] However, if the hold-harmless level were less than 100 percent and 
poverty data remained stable, funding would eventually fully adjust to 
the distribution of low-income children. 

[11] The appropriations’ provisions entitle a state to receive 100 
percent of its previous year’s funding regardless of the number of 
districts in that state that meet the minimum eligibility requirements. 
In cases where there are individual districts within a state that no 
longer meet the eligibility requirements for a particular grant, the 
state’s full grant amount gets distributed among the qualifying 
districts for that year. 

[12] If the total appropriation for the program is not large enough to 
fully fund the larger of these two amounts, then every district’s 
allotment is reduced proportionally. This allows more of the Title I 
appropriation to be distributed on the basis of the poverty criteria in 
the formula. 

[13] Districts with fewer than 1,000 children enrolled are exempt from 
this requirement. 

[14] As long as the districts use the same measure of poverty across 
all schools, they can choose to use free and reduced-price lunch, 
Medicaid, TANF, or other measures. The measure used by the district 
does not have to be the same measure as that used by the state or the 
federal government in allocating funds. 

[15] In this report, analyses of numbers of poor children at the school 
level include only children in public schools. 

[16] This relationship holds true whether poverty is measured using 
Census data or the number of children receiving free or reduced-price 
lunch. The subsidized lunch program provides a looser definition of 
“poverty” than the Census poverty data. Eligibility for free lunches is 
set at 130 percent of the official poverty line ($22,165 for a family 
of four during the 2000-01 school year), and eligibility for reduced-
price lunches extends up to 185 percent of the poverty line ($31,543 
for a family of four during the 2000-01 school year). The number of 
students eligible for subsidized lunches is roughly double the number 
meeting the Census poverty definition. Nonetheless, according to the 
Department of Education, the subsidized lunch program provides the best 
available source of data on low-income students at the school level. 

[17] We examined this issue using basic grants alone, rather than 
combining basic and concentration grants, because many districts that 
receive basic grants do not receive concentration grants and should not 
be included in a state-level analysis with those that receive both 
grants. Nonetheless, an examination of concentration grants alone 
showed that they, too, were unevenly distributed across states and 
eligible districts - a point that is discussed later in this report. 

[18] Poor children in these analyses are children eligible for free or 
reduced-price lunch. We used subsidized lunch eligibility for this 
analysis because these data are available for a greater number of 
school districts than are Census data. 

[19] In these examples, poverty was measured as eligibility for free or 
reduced-price lunch. 

[20] In general, there is a strong relationship at the district level 
between the number of children eligible for free and reduced-price 
lunch and the number of children denoted by the Census as living in 
poverty. However, among urban districts that relationship is closer to 
a one-to-one relationship than in rural districts. In rural districts, 
the relationship approaches a two-to-one relationship, that is two 
children are identified as eligible for free or reduced-price lunch for 
every one child denoted by the Census as living in poverty. 

[21] Education classifies school districts into eight location types. 
These are: (1) large central city, (2) mid-size central city, (3) urban 
fringe of large city, (4) urban fringe of mid-size city, (5) large 
town, (6) small town, (7) rural-outside a metropolitan area, and (8) 
rural-inside a metropolitan area. In our analysis, classes (1) and (2) 
are considered “urban,” and classes (7) and (8) are considered “rural.” 

[22] The enrollments of the charter schools ranged to a high of 2,099 
students. Therefore, actual allocations to charter schools were 
compared with those for other school districts with fewer than 2,500 
students. Census poverty data are unavailable for charter schools, so 
all analysis used free or reduced-price lunch eligibility as the 
measure of poverty. Charter schools that are independent school 
districts in Wisconsin and Illinois were not considered in the analysis 
as the three such charter schools in Illinois were not funded in 1999-
2000, and only one of the two such charter schools in Wisconsin was 
funded. 

[23] Along with the District of Columbia, the 14 states were: Arizona, 
California, Delaware, Illinois, Louisiana, Massachusetts, Michigan, 
Minnesota, New Jersey, North Carolina, Ohio, Pennsylvania, Texas, and 
Wisconsin. 

[24] In addition, districts must fund all schools with poverty rates 
greater than 75 percent before funding any school with a poverty rate 
less than 75 percent. 

[25] In responding to our survey, school district officials self-
identified their districts as urban, rural, suburban, or mixed. 

[26] Confidence intervals for these estimates are provided in appendix 
I. 

[27] For some districts, the U.S. Department of Education approves 
these waivers directly; however, under the “Ed-Flex” program, Education 
has delegated this authority to 12 state education agencies. 

[28] This shift could not be examined at the school district level 
because Education did not generate school district-level allocation 
amounts until the 1999-2000 school year. 

[29] The effect of hold-harmless is important if the amount of Title I 
funds remain the same, but if funds are increased the impact of the 
hold-harmless is limited. 

[30] This analysis holds constant at the 2001-02 levels, the total 
funding, the hold-harmless rule and the per-pupil expenditure factor. 

[31] The definition of poor child here is the poverty factor used in 
the formula that includes Foster, Neglected, TANF, and Delinquent 
children with the Census poverty estimates. 

[32] School districts in states with growing poverty that were also 
recipients of the small state minimum are nevertheless much better off. 
While their grant amount per poor child would have fallen (due to a 
proportionally more rapid increase in their number of poor children 
than in the increase in their grant amount), the amount per poor child 
of such states still exceeds that of states not benefiting from the 
small state minimum. 

[33] For purposes of comparison, the average grant per poor child in 
2001 is $679 for basic grants and $149 for concentration grants. 

[34] School Finance: State and Federal Efforts to Target Poor Students 
[hyperlink, http://www.gao.gov/products/GAO/HEHS-98-36], Jan. 28, 1998. 

[35] For this analysis, school districts were rank-ordered on the basis 
of their numbers of low-income children and divided into five groups, 
or quintiles, each containing approximately 20 percent of all low-
income children. Districts with the smallest number of children were in 
the first group and districts with the greatest number of children were 
in the fifth group. The size and number of the school districts in each 
group differ. For example, the school districts in the first group have 
numbers of poor children that range from 1 to 643 and the group 
contains nearly half of all districts, while the school districts in 
the fifth group have numbers of such children ranging from 26,976 to 
303,122 and this group contains 6 percent of all districts. Also, 
differences in funding among districts within each of the five groups 
are measured by the standard deviation of the funding per child. 

[36] This represents a 2-year change in the number of low-income 
children. School year 2000-01 used poverty estimates for calendar year 
1995, and school year 2001-02 used estimates for 1997. 

[37] Using a less restrictive hold-harmless rule could also increase 
the volatility of school district funding, especially among small 
districts for which Census estimates of poor children contain 
substantial statistical error. 

[38] Studies of state spending on education (and most other public 
services as well) consistently show that education spending is strongly 
related to the income of state and local taxpayers. For this reason, 
the use of per-pupil expenditures as an indicator of differences in the 
cost of delivering educational services has been severely criticized in 
the past. 

[39] W. Riddle and R. Apling, Education for the Disadvantaged: 
Allocation Formula Issues in ESEA Title I Reauthorization Legislation, 
Congressional Research Service #RL-30492, March 20, 2000, pages 14-17. 
For a general discussion of educational costs and expenditures, see 
U.S. Department of Education, National Center for Education Statistics, 
A Primer for Making Cost Adjustments in Education, NCES 2001–323, by 
W.J. Fowler, Jr. and D.H. Monk. 

[40] See Geographical Variations in Public Schools’ Costs, National 
Center For Education Statistics, Working Paper No. 98-04, February 1998 
for a description of the cost factor and its methodology. We used 
estimates for school year 1993-94, the latest available at the time of 
this report. 

[41] For example, cost-of-living and other “amenity” factors, such as 
climate, geography, and area crime rates, could result in differing 
teacher salaries when these factors differ significantly among school 
districts. The inclusion of “amenity” factors represents a departure 
from earlier experimental models that focused primarily on differences 
in teacher experience and education. While controlling for amenity 
factors that may affect the salaries necessary to attract teachers to 
specific locations is, in principle, appropriate, the measurement 
methodology behind these attempts is subject to a variety of 
statistical specifications that could lead to varying results. In 
addition, we did not investigate the quality of the data used in these 
models. For these reasons, we have labeled these estimates as 
experimental and do not endorse any particular approach. 

[42] The effect of increased targeting to districts with high 
percentages of poor children occurs because the correlation between per-
pupil spending and the percentage of poor children and the correlation 
between the cost factor and the percentage of poor children differ. 
Because these correlations can change from one year to the next, the 
targeting pattern in future years may differ from that shown here. 

[43] To qualify, a district must have a poverty rate of more than 2 
percent and at least 10 children in poverty. 

[44] Formula-eligible children are children living in poverty as 
reported by the Bureau of the Census, plus Foster Children, TANF 
children, Neglected Children, and Delinquent Children as reported by 
the Department of Health and Human Services. 

[45] Public Law 107-110. 

[46] Public Law 107-116. 

[End of section] 

GAO’s Mission: 

The General Accounting Office, the investigative arm of Congress, 
exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO’s commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO’s Web site [hyperlink, 
http://www.gao.gov] contains abstracts and full text files of current 
reports and testimony and an expanding archive of older products. The 
Web site features a search engine to help you locate documents using 
key words and phrases. You can print these documents in their entirety, 
including charts and other graphics. 

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as “Today’s Reports,” on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
[hyperlink, http://www.gao.gov] and select “Subscribe to daily E-mail 
alert for newly released products” under the GAO Reports heading. 

Order by Mail or Phone: 

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 

Orders should be sent to: 

U.S. General Accounting Office: 
441 G Street NW, Room LM: 
Washington, D.C. 20548: 

To order by Phone: 
Voice: (202) 512-6000: 
TDD: (202) 512-2537: 
Fax: (202) 512-6061: 

To Report Fraud, Waste, and Abuse in Federal Programs Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 

E-mail: fraudnet@gao.gov: 

Automated answering system: (800) 424-5454 or (202) 512-7470: 

Public Affairs: 
Jeff Nelligan, managing director, NelliganJ@gao.gov: 
(202) 512-4800: 
U.S. General Accounting Office: 
441 G Street NW, Room 7149:
Washington, D.C. 20548: