June 28, 2007 | Special Report on Education
Since its inception in 1965, the Elementary and Secondary Education Act, now known as No Child Left Behind (NCLB), has directed billions of federal dollars toward low-income students. Title I, Part A of NCLB is designed to equalize educational opportunities and resources for disadvantaged children.
This analysis examines whether the current mechanisms for providing federal education funding to disadvantaged children are effective and whether the system works as originally intended. The evidence yields the following major findings.
A History of the ESEA and Title I
As part of the "War on Poverty," the Elementary and Secondary Education (ESEA) Act of 1965 was designed to direct federal education dollars to the most disadvantaged children living in poverty. Consequently, this landmark legislation was undertaken specifically to authorize the federal government to equalize the educational opportunities of all children. In addition to creating a federal role in directing public education dollars to policy goals, such as eliminating poverty, the ESEA was designed to transfer funding through state governments, thereby resulting in substantial increases in education bureaucracy at the state level.
Since 1965, the ESEA has been reauthorized eight times. Among these reauthorizations have been the Improving America's School Act (IASA) of 1994 and, most recently, the No Child Left Behind (NCLB) Act of 2002. With each transition, the funding mechanisms have become more complicated, and the bureaucracy needed to implement the program has grown.
The U.S. Department of Education and numerous other researchers have sought continually to assess the effectiveness of the federal role in education in terms of its impact on the academic achievement of the targeted populations, but much less attention has been given to the effectiveness of the funding approach. In other words, how well does the current system of federal compensatory education achieve its original purpose? Moreover, have the funding formula changes that have been made over the past four decades helped or hindered the effectiveness of Title I?
Title I, Part A of NCLB specifically addresses federal compensatory education for disadvantaged children by determining which students are eligible and, in theory, how much they are eligible to receive. This analysis examines the current format of Title I of NCLB with a focus on its increasing complexity and the implications of that complexity.
It is important to understand how funds appropriated by Congress under NCLB actually flow to the targeted students, since significant discretion at both the state and local levels results in variations in amounts per student between the Local Education Associations (LEAs) and schools that have similar demographics. In addition, the effectiveness of Title I, Part A in reaching its targeted population is a critical point to be considered. With a cost of nearly $13 billion in federal funds, is this program achieving its intended purpose, or might there be a better way to direct resources to those students who are the intended beneficiaries?
Compensatory Education Funding: Increasingly Complex and Obscure
In general, public education funding in the U.S. over the past century has evolved from a fairly straightforward system of local residents contributing a portion of their income (usually based on the amount of property that they own) to the education of the community's children into a complex system of state and federal funding formulas. In 1920, over 80 percent of the revenue for public elementary and secondary schools was generated at the local level. By the 2001-2002 school year, that number had dropped to less than 43 percent.
More important, the funding mechanisms have become so complex at the state level that only a handful of people in each state can claim to understand them. As a result, it is nearly impossible to determine how a school's total funding changes as children enter or leave their public school systems.
Similarly, the federal government's funding for compensatory education programs has grown from a single grant program that assigned a roughly similar supplementary amount to every eligible child to a set of four major grant programs and a growing list of categorical programs. Each additional formula grant, moreover, has been more complex than the previous one.
Thus, although a child living below poverty in the United States is clearly eligible under Title I to receive some amount of federal compensatory funding for his or her education, determining the particular amount that child actually receives is nearly impossible. While funding increases lead to more money in the system, they flow into a confusing distribution maze, and it is difficult to tell what additional benefit, if any, they provide to individual students. This calls into question the efficacy of further funding increases.
How Title I Works: Four Major Grant Programs
As noted, Part A of Title I currently contains four separate grants for disadvantaged students. The main program, begun at the inception of ESEA in 1965, is the Basic Grant. In FY 2005, the total Basic Grant appropriation was $6.9 billion, or 57 percent of the total Title I, Part A appropriation. Shifting priorities since that time have resulted in the Basic Grant appropriation's remaining steady at approximately $7 billion, while the other three grants-Concentration, Targeted, and Education Finance Incentive Grants-now comprise 46 percent of the over $12.7 billion appropriated to Title I, Part A in FY 2006.
The Basic Grant is the least complex of the four types. The total amount of a Basic Grant allocated to a state, referred to in the legislation as a State Education Agency, or SEA, is determined by multiplying the total number of eligible children in each of the state's Local Education Associations by a dollar amount that is 40 percent of the state per-pupil expenditure (SPPE), provided that this amount falls between 32 percent and 48 percent of the national average SPPE.
The SPPE was included in the formula to capture differences in the cost of education between states. However, it has come under considerable criticism both for being a uniform number for an entire state, regardless of urban, rural, or other potential cost of living differences, and for representing a state's willingness to fund education rather than simply a cost difference.
Eligible children are defined as those between the ages of five and 17 living below the federal poverty threshold in each LEA, regardless of whether they attend public schools. The count also includes children in foster homes, children living above poverty but receiving Temporary Assistance to Needy Families (TANF), and children in institutions for neglected and delinquent children. Until recently, the U.S. Census Bureau released the eligible children count every 10 years, but it is now based on an annually updated number, the Small Area Income and Poverty Estimates (SAIPE). This number, however, is usually several years out of date.
To participate in the Basic Grant program, an LEA must have at least 10 eligible children who represent more than 2 percent of the school-age population in the LEA. For the 2003-2004 school year, approximately 150 of the over 14,000 LEAs in the U.S. (only 0.01 percent) did not meet these criteria. It does not follow from this statistic, however, that over 99 percent of LEAs received Title I funding, as will be explained in greater detail later.
The result of the formula (SPPE times the number of eligible children) should determine an LEA's Basic Grant allotment; in reality, all it initially represents is the relative proportion of that LEA's needs compared to the total funds allotted for all LEAs. The authorized levels are higher than appropriations to date. As a result, each LEA's allotment is ratably reduced by multiplying its proportion of the total allotment by Congress's total appropriation. Once this initial allocation has been determined, however, further calculations must be made to determine state and local grants.
Some of the most frequently criticized provisions of the federal grant program are its hold-harmless provisions. For example, districts with particular percentages of poverty rates are guaranteed certain funding the next year, despite any decrease in poverty that they might experience. Districts with up to a 15 percent poverty rate are guaranteed to receive at least 85 percent of their prior year's funding; districts with 15 percent to 30 percent poverty are guaranteed at least 90 percent; and districts with more than 35 percent are guaranteed at least 95 percent of the prior year's grant.
Once the hold-harmless amounts are compared to current-year allocations, funds are ratably shifted from districts that are not affected by hold-harmless to districts that are affected. As a result, by retaining rates of guaranteed funding, districts with declining poverty can siphon funds from districts with growing poverty. Of course, this process may need to be performed multiple times if the ratable reduction causes an initially unaffected district to become affected.
Finally, the total Basic Grant allocation for any state cannot be less than either 0.25 percent of the total U.S. allocation or the number of eligible children times 150 percent of the national average per-pupil payment. This means that small states, such as Vermont, New Hampshire, Wyoming, or Alaska, receive a much larger amount per child than larger states, regardless of socioeconomic status. After the small-state allocation of funds, the ratable reductions must be repeated.
Utah: An Example of How a Basic Grant Is Awarded. Using the state of Utah as an example, the grant process can be demonstrated up to the point of hold harmless, as seen in Table 1. To determine the 2003-2004 Basic Grant allocation, Utah's state per pupil expenditure of $5,008 is used. Forty percent of Utah's SPPE is $2,003, and because this amount is less than 32 percent of the U.S. SPPE average ($2,620), the federal government would use the minimum U.S. SPPE to determine the initial Basic Grant funding.
According to the U.S. Census Bureau, there are 49,493 eligible children in Utah. Multiplying the number of eligible children by the minimum SPPE yields an initial amount for Utah of $129.7 million. However, because the federal government allotted only $6.8 billion for the 2003-2004 school year, the initial amounts appropriated to each of Utah's 40 LEAs are compared to the other 14,250 districts in the nation in order to ratably reduce the state's allocation to $33.7 million. At this point, the hold-harmless comparisons would be made. Beyond that, Utah appears to have 0.5 percent of the total spent on Title I throughout the nation, making it unnecessary to apply the small-state provision. However, the amount calculated for each of Utah's 40 districts could be reduced once the small-state provision is applied to qualifying states.
Similar processes would then be carried out for the other three Title I, Part A grants-Concentration Grants, Targeted Grants, and Education Finance Incentive Grants. The total Title I, Part A funds granted to Utah in 2003-2004 was $50.8 million. Because Utah has only 40 LEAs, computations are not as extensive as they are for other states. For example, California has nearly 1,000 school districts, which require much more calculation. Regardless of the complexity of other states' districting, Table 1 demonstrates that all LEAs in a given state use the same SPPE amount, whether they are urban, suburban, rural, high-poverty, or low-poverty.
To supplement the Basic Grant, the Concentration Grant was added in 1978 as part of an effort to focus funds on districts with higher numbers of low-income students. It is calculated in same manner as the Basic Grant, with two exceptions.
First, to be eligible to receive this grant, an LEA must have at least 6,500 eligible students, or else 15 percent of the total number of students must be eligible.
Second, the hold-harmless provision for the Concentration Grant guarantees prior-year funding for four years instead of one, even if the LEA is no longer deemed eligible for the grant.
Targeted and Education Finance Incentive Grants
In 1994, when the ESEA became the Improving America's School Act (IASA), two more grant categories were added: the Targeted Grant and the Education Finance Incentive Grant (EFIG). The purpose of these two grants was, like the purpose of the Concentration Grant, to direct funds to those districts with the greatest percentage of children living in poverty. LEAs are eligible for both the Targeted Grant and the EFIG if they have more than 10 eligible students and this number is greater than 5 percent of the total number of children in the LEA. These thresholds are slightly different from those for either Basic or Concentration Grants.
For the Targeted Grant, eligible children are weighted by factors ranging from 1.0 to 4.0, based on any of the four thresholds: the percent or number of qualifying children in each LEA or county. There are five weighting categories for each of the four types of thresholds, and there are different weights for the percent calculations versus the number-of-children calculations. The lowest weightings are for LEAs with less than 15.58 percent poverty, or 691 children, and for counties with less than 15 percent poverty, or 2,311 children. The highest categories are for LEAs with more than 38.24 percent poverty, or 35,514 children, and for counties with more than 29.2 percent poverty, or 93,811 children. After the different categories are summed for each qualifying LEA or county, the higher number is used.
The rationale behind the particular thresholds is not entirely clear in the legislation. However, it would seem to suggest a level of precision in the Census data that may not be practical.
While the amount of information required for Targeted Grants is substantially more than is required for either Basic or Concentration Grants, it pales in comparison to the data required for Education Finance Incentive Grants. First, it should be noted, the SPPE used for the EFIG is restricted to between 34 percent and 46 percent of the national average rather than the 32 percent to 48 percent range of the other three grants. Again, these limitations simply affect the proportional amount ascribed to each LEA prior to ratable-reduction, hold-harmless, and small-state provisions. Nonetheless, using different amounts for different grants adds to the program's complexity and red tape.
Similar to the Targeted Grant, the EFIG requires a weighted student count. However, for the EFIG, the count is multiplied by an "effort factor," which represents the ratio of the three-year average SPPE for a given state multiplied by the three-year average per capita income in the United States to the three-year average per capita income in each state multiplied by the three-year average SPPE for the U.S. In other words, each state's "effort" is how much of its per-capita income it devotes to public education as compared to the national average.
In addition, an "equity factor" is calculated for each state, based on a weighted coefficient of variation. This number represents a measure of variation in per-pupil expenditures across a state. The EFIG implicitly assumes that equal total per-pupil spending across districts and states is the ideal, though the effectiveness of funding equalization has been debated. States are then divided into those with an equity factor of less than 0.10, those between 0.10 and 0.20, and those with an equity factor of greater than 0.20. Each of these groups then has unique student weighting rules similar to those of the Targeted Grant-five categories of weights for percentages and number of children for both LEAs and counties. In other words, there are 60 weighting categories for the EFIG rather than the merely 20 used for the Targeted Grant.
Clearly, determining the amount of federal aid generated by a particular student living below poverty is extremely complex. Nonetheless, once the four grant categories have been calculated, adjusted, and recalculated for each LEA or county, they are summed for each of the 50 states. This amount is that state's Title I, Part A allocation. As a result of this complicated process, the underlying calculations are not well understood, either by those administering the program or by the public in general.
After the federal government appropriates the funds to states, the focus then shifts to distributing the total state allocations, both to the eligible LEAs and then to the schools within each LEA, ideally resulting in some measurable classroom impact on the students living below poverty. There are two important factors in this process, however: first, the complexity of the guidelines for the SEA and the LEAs in distributing funds and, second, the monetary amounts that can be retained at each level, either for administration or for particular programs.
Distributing Federal Compensatory Funds: Dilution and Discretion
Each state's total Title I, Part A allocations are calculated by adding together the four grant categories for every LEA in the state. Although the U.S. Department of Education informs the LEAs that they are to receive funds, it is the state that distributes the funds based on a separate set of guidelines.
Before distributing funds, the state reserves a portion of the total for various tasks. In terms of set-asides, states may reserve up to 1 percent or $400,000 for administrative duties. In addition, states must reserve 4 percent of the amount received for school improvement. Of this amount, at least 95 percent must be directed at schools identified as in need of improvement. The funds may be used for school support and for the state to recognize academic achievement gains made by LEAs.
Technically, a district's Title I, Part A allocation should be approximately 95 percent of the amount calculated by the National Center for Education Statistics (NCES) for the four formula grant programs. However, this is often not the case, as states have to redistribute funds to account for new districts or charter schools that are not contained in the NCES database. States can also use alternative poverty data, such as the number of children that qualify for free or reduced-price lunch, to distribute funds.
District size affects Title I distribution even further. Any districts with fewer than 20,000 total residents are considered "small LEAs" and are not subject to the NCES grant calculations. States can redistribute the total amount calculated for small LEAs based on "population data that the State educational agency determines best reflects the current distribution of children in poor families."
This small LEA designation applies to nearly 80 percent of public school districts. For this reason, tying the data on total Title I, Part A allocations as distributed by the U.S. Department of Education to data reported by states or districts is usually problematic.
Finally, before distributing the remaining Title I funds to schools, districts must also set aside funds for specific programs. Under NCLB:
This means that in some cases, only 64 percent of an LEA's remaining Title I allocation is available to be distributed to schools for general use. If these set-asides were not mandated, states and LEAs would have the ability to use their Title I funds in ways they believe to be most effective.
Once the funds have been distributed to the LEAs, they must then be directed to the eligible schools. Federal guidelines require that LEAs first serve those schools with more than 75 percent of their students living in poverty. Most schools rely on free and reduced-priced lunch data to determine eligibility.
Then, after all schools meeting this threshold have been served, districts rank order their schools according to the percentage of low-income students. This can be done with the Census data, the free and reduced-price lunch data, TANF data, Medicaid data, or a composite of all of these. Districts serve the schools in order of decreasing rates, down to those that have at least 35 percent eligible students. To help LEAs determine how to allocate funds, the U.S. Department of Education has published a Non-Regulatory Guidance. The guidelines are very specific and somewhat complicated.
Table 2, taken from the Non-Regulatory Guidance, demonstrates how LEAs can follow department guidance by serving schools down to the level of 35 percent poverty; it also demonstrates the complexity of distribution. This table is one of several examples that LEAs can use for guidance. Ultimately, however, the school board and superintendent are responsible for determining the strategy for the allocation of funds. It is also their responsibility to ensure compliance with the associated federal guidelines.
Florida: An Example of Title I Distributions. According to the U.S. Department of Education, 67 Florida districts were to receive a total of $639.2 million during the 2005-2006 school year. Of this amount, the Florida Department of Education distributed $573.1 million to its LEAs, with over $67 million, or 10.5 percent, retained by the state. Over one-third of the $573.1 million was then set aside for the categorical programs designated in the legislation. As Table 2 demonstrates, Florida's state-level discretion allowed approximately $367 million, or 57 percent of the total federal allocation, to be distributed to the schools.
Once the LEAs are granted authority to distribute funds, it appears that Florida districts use variable criteria to determine school-level allocations. Some districts, such as Hillsborough, distribute a flat amount of $500 for each child who is eligible to receive free and reduced-price lunch, regardless of total school poverty or grade level. Other districts, such as Palm Beach, appear to distribute $259 per student for schools with 50 percent to 60 percent eligibility, $324 for schools that are between 61 percent and 75 percent eligible, $389 for schools that are between 75 percent and 90 percent eligible, and $486 for schools with more than 90 percent of their students eligible for free and reduced-price lunch.
In fact, there does not appear to be any discernible pattern in the amounts of school-level distributions in Florida. The smallest amount distributed per student was $142 in Putnam County, and the largest amount was $1,044 in St. John's County.
Title I, Part A Funds: Results vs. Intentions
Having shown how Title I, Part A funds are calculated for each state and how they are to be distributed to districts and schools, it is necessary to examine whether the current Title I system is the most appropriate way to provide compensatory education funding to assist disadvantaged students. The system's complexity and variability make it difficult for the public to determine how well the federal government is directing resources to the intended beneficiaries.
At a minimum, a state's Title I allocation should have some relationship to the number of students living below poverty in the state. To account for the differences in cost of living between states, the federal government uses the state per-pupil expenditure. To prevent great variability, the government then limits the SPPE differences between 32 percent and 48 percent of the U.S. average.
However, Chart 1 suggests that there is much disparity between the states without regard for need. Students in some of the less populated states, such as Vermont or Wyoming, receive, on average, over $3,000 per child. Quite a few of the more populated states, on the other hand, receive less than $1,200 per child. As a result of these inconsistencies, some states receive 250 percent of the amount per pupil that other states receive.
The greatest causes of this variability are the hold-harmless and small-state provisions. Consequently, some states with a higher number of disadvantaged students, such as Kentucky, Mississippi, or Missouri, actually receive substantially less per child than others receive. In addition, when poverty shifts from one state to another, the funds take several years to follow.
For example, the Title I program apparently is intended to focus funds where the greatest concentrations of poverty levels are located. Thus, there should be higher funding per student in states with higher percentages of children living in poverty.
However, Chart 1 disputes this contention as well. New Hampshire has one of the lowest rates of poverty at 5.2 percent and one of the highest amounts of Title I funds per student: $2,294. Arkansas is just the opposite: Nearly 22 percent of its children are living in poverty, while the per-student amount is only $1,185. One can argue the merits of the concentration of poverty approach, but the law does not appear to be having the effect that the law intended, at least at the state level.
In addition to the great disparity at the federal and state levels, districts are given discretion over the distribution of their federal funds. The intention here is to allow districts the flexibility to use funds in the most effective way for their given set of circumstances. It would seem that more funds would be directed at schools with higher percentages of low-income students, or at least at schools with the lowest performance.
However, Chart 2 suggests no clear pattern in Florida. From a concentration of poverty perspective, the highest per-pupil school allocation is for schools with between 70 percent and 80 percent of students qualifying for free or reduced-price lunch, not the highest levels of poverty. More interesting is that the pattern of per-pupil amounts by school performance is counterintuitive. Florida administers a yearly standardized test, the FCAT, and grades schools based on the results. Schools with the lowest grade, "F," had the lowest average per-pupil amount of federal funds, while the "B" schools had the highest.
What Congress Should Do
Because American taxpayers spend approximately $13 billion on the Title I program-and hundreds of billions of dollars on K-12 public education all told-it is important that education funding be spent in an appropriate manner and that policies are designed to minimize waste, fraud, and abuse. Excessive complexity only opens the door to error and makes mismanagement more easily hidden.
Yet Congress, over many decades, has continued to add layer upon layer to Title I so that it is implemented in a manner that is different from the original legislative intention. Moreover, the program's complexity increases the challenge of fiscal accountability and allows only a few people to understand the process.
For example, the current system of formula grants has become increasingly difficult to understand and implement, and this in turn has augmented the bureaucratic and administrative costs of implementation. The result has been that less funding reaches the program's intended beneficiaries: disadvantaged students and the schools they attend. Moreover, this complexity has led to great variance in the amount of funding that is provided to local communities across the country relative to their population of disadvantaged students, and the system's increasing bureaucratic complexity has clouded transparency and public understanding, thereby undermining accountability for the largest federal K-12 program.
Title I should therefore be reformed to achieve greater transparency and a more student-centered approach. Specifically, Congress should:
In the course of eight reauthorizations, Title I has accumulated great bureaucratic complexity. The program's funding formulas have become increasingly complex and obscure.
Today, the distribution of funds through the Title I program is characterized by unintended variability. There is little or no relationship between a district's demographics and the amount of money received for compensatory education aid assistance. Moreover, because the current funding formula reserves too much funding for administrative costs, less money is available for instructional expenses.
Congress should (1) reform the Title I program to return to the original student-centered goal of compensatory education; (2) streamline the funding formula to use a simple per-pupil allocation to improve transparency; and (3) allow states to implement simplified student-centered funding that provides aid more directly to students, thereby allowing greater portability and school choice.
Susan L. Aud, Ph.D., is a Senior Fellow with the Milton and Rose D. Friedman Foundation and teaches doctoral-level quantitative methods courses as a professorial lecturer at the Paul H. Nitze School for Advanced International Studies at Johns Hopkins University. Her research has been focused predominately on K-12 public education fiscal policy, as well as the competitive and efficiency effects of school choice on the public school system. Dr. Aud received her Ph.D. from George Mason University and her MBA in finance from George Washington University.