Sections

Research

What Title I portability would mean for the distribution of federal education aid

Penelope Ghartey plays a game on his phone at his home in Brooklyn, New York

Executive summary

Congressional attempts in the last few years to allow states to divert their Title I funds through the Elementary and Secondary Education Act into vouchers that would follow individual poor students to the public or private schools of their choice—Title I “portability”—were unsuccessful. Such vouchers are now a far more politically realistic possibility. It seems most likely that policy choices around vouchers would be made at the state level. While this reassured many senators during Secretary DeVos’ confirmation process, it may not reassure those at the local level.

Current Title I rules allow school districts to concentrate their federal funds in their poorest schools, while under a portability state option, school districts would be forced to distribute those funds uniformly per poor pupil. This shift in allocation rules means that portability would affect the distribution of funds not only across districts, but also within them—and that this shift would happen among public schools even in districts where few, if any, students chose to take Title I funds to private schools.


Introduction

The unprecedented public interest in the DeVos nomination reflects great interest in how Congress and the Trump administration might try to promote school choice via the federal role. The administration has not offered a school choice policy proposal with any specific details to date, and education policy watchers are looking to past proposals and state level policies. One relatively fleshed-out policy proposal in the mix is the concept of Title I “portability,” proposed in the House and Senate prior to the passage of the Every Student Succeeds Act (ESSA). The basic concept of making federal funds “portable” is often explained with the analogy of putting the federal cash into a student’s backpack that moves with the student to whatever school he or she attends, public or private. As with any school finance policy, however, the specific details of any portability proposal would be critical. Absent an unlikely considerable influx of new federal funds, portability would fill students’ backpacks using at least in part cash that used to be appropriated to school districts through Title I funds.

Much of the policy discussion focuses on the highly salient fact that these vouchers could be used in private schools, prompting many questions. Would vouchers on the order of $1,400 per pupil allow poor students access to private schools? Would poor students using vouchers to attend private schools do better than if they remained in their public systems? Would the competitive pressure from private schools improve public schools in the absence of resource changes? Are public schools spending Title I funds effectively now, and what would change in their absence? The evidence base for these questions is emerging, and answers often depend on local context.

These are critical questions, but this discussion of portability misses an important point. Any voucher-type proposal that relies on funds currently appropriated to Title I would not simply shift funds from public to private schools—it would significantly redistribute federal funds within states across school districts, and within districts across public schools, reducing progressivity. This redistribution would spread federal dollars more uniformly across schools rather than concentrating funds in the highest poverty schools within districts, and districts within states, even if few students chose to take their federal dollars to private schools. Districts with few private schools could still feel the impact of portability if their states choose to take it up.

In this report, I focus on the short-run fiscal impact of portability, showing how it would affect the distribution of federal funds across school districts within a state, and across schools within a district. I describe a hypothetical portability policy quite similar to that proposed by members of Congress in recent years, in which state-level Title I allocations would remain the same as in the current regime. I use Maryland as an example due to its small number of school districts.

The big picture: Should local conditions affect compensatory grants per poor pupil?

If no students took their backpacks of cash to private school, the difference between Title I and the recent portability proposals boils down to whether you think poor students should get the same amount of federal funds no matter where they live and go to school. There are theoretical arguments to be made on both sides of this issue.

The current research consensus points to high-poverty schools facing disproportionate challenges: that is, to get the same outcome, they need more resources per student—and per poor student. This would argue for progressive funding based on the poverty level at the school level.

However, any formula that offers higher-poverty schools—or districts—greater funding per poor pupil provides an incentive for economic segregation of students in order to maximize federal funding. This is particularly relevant at the school building level, as school district policies—and other local government decisions about zoning, development, and transportation—can affect how economically segregated a district’s public schools are. Just last year, parents in Loudoun County Public Schools in Virginia argued—ultimately unsuccessfully—to revise school attendance boundaries to make more segregated schools. Proponents of the segregation plan argue that the creation of higher needs schools would generate more funds for disadvantaged students.

A flat per-poor-pupil grant (such as through portability) would neither incentivize economic segregation, nor provide high-poverty schools additional funds for meeting their disproportionate needs; Title I does both. Weighing these tradeoffs requires some assumption about how responsive parties will be to incentives to segregate. In this piece, I provide an analysis of the short-run partial equilibrium effects of portability on school-level Title I revenues. I assume that no students would move within or between districts as a result of portability, and assess how it would change the amount of Title I revenue per poor student enrolled in a school.

How Title I redistributes federal funds now

Because portability yields a flat distribution of funds per poor student, the impact of the change would be a direct function of current funding mechanisms. While the Every Student Succeeds Act of 2015 made many changes to its predecessor, the No Child Left Behind Act, it did not change the formulas for Title I Part A. These formulas send some funds on a uniform, per-disadvantaged-child basis, but direct close to half of federal dollars in a way that recognizes high-poverty districts face greater challenges, using weights to allocate per-eligible funds progressively with respect to a district’s poverty rate. These weights mean two states with the same number of poor and non-poor students would get more Title I funds if those students are economically segregated across school districts. District decisions about how to distribute Title I funds across their schools do not change the total district allocation.

Different districts, even within a given state, wind up with highly variable amounts of Title I funds per eligible child. Some aspects of this variance are visible in the design of the four formulas, while others are less transparent. The formulas are complex and opaque: in the Every Student Succeeds Act, Congress asked the Institute of Education Sciences to issue a report explaining how the current formulas affect different types of districts and school attendance areas.

More details on Title I formulas

Title I Part A has four formulas that send different funding levels per Title I-eligible (essentially, poor) child depending on state education spending levels. Basic grants send the same amount per eligible child to each district within a state, provided the district meets very minimal standards (at ten poor children and at least two percent child poverty rate). Concentration grants do the same, but only for districts with a higher minimum threshold child poverty population (over 6,500) or rate (at least 15 percent). Targeted grants are designed to send more funds per poor student to districts with greater rates or counts of child poverty. And the Education Finance Incentive Grants are an attempt to make states spend more, and more equally, on their schools.

 

The four formulas also have their own hold-harmless provisions, which, in combination with appropriations that have fallen short of full funding in each year after the program’s first, 1965, mean that historical funding levels continue to influence current ones. Two districts with similar demographics can have different Title I allocations because of differences in their past funding levels. And small-state minimum provisions in the formulas guarantee minimum allocations to each state, no matter how few funds they would otherwise qualify for—this generates visibly higher allocations for low population states.


How Title I could be converted to portable vouchers

Portability has been proposed as a state option under Title I. Total state-level Title I allocations from the federal government would not change regardless of whether a state chose the portability option. But any state choosing this option would experience changes in how Title I funds get divided among and within its school districts—even if all poor students were to attend public schools. Proponents of portability like this feature of it: it switches the focus of policy from what schools get to what individual children get.

For example, a high-poverty district with lots of Title I-eligible (poor) students weighted heavily in the targeted grants formula would continue to generate disproportionate federal funds for its state’s total allocation. But under portability, those funds would then be allocated down to school districts within the state uniformly, disregarding concentrations of poverty. The state would divide its total Title I allocation by its number of eligible (poor) children, and that would be the voucher amount following each poor student to his or her school of choice—public or private. Under the portability option, the per-eligible grant amount would be the same in every school district within a state.

Within school districts, Title I funds would follow each poor student to the school he or she attends, public or private. The distinction between Title I and non-Title I schools would vanish, as any school enrolling even one poor student would receive Title I funds.

Assumptions about portability policy for simulations

I simulate the effects of using all Title I funds appropriated in FY2015 for vouchers to Title I-eligible (poor) school-aged children. I assume that the Title I formulas continue to determine state-level allocations, and that these state Title I pies are then divided equally among all poor children in the state.

Were Congress to enact some semblance of portability, Congress—and any states choosing to take up the option—would need to address questions like under what conditions (tuition, selectivity, compliance with state and federal curricular and civil rights requirements) private schools would be allowed to accept the vouchers. Because my interest here is in how portability would affect the distribution of federal funds across public schools even if students do not switch to private schools, I do not need any assumptions about the terms private schools would face.

Portability simulations

The Education Trust and the Center for American Progress released reports last year simulating district-level federal funds under portability. These reports show how portability would transfer funds within states from districts with higher poverty rates to less poor districts. My results are consistent with those findings. It is my goal here to show how within an individual state (where, as most recently proposed, portability’s fiscal impact would be), portability would change the distribution of Title I funds across all districts, and within a district, across schools currently participating in Title I versus those who do not. I use Maryland for this exercise because it has relatively few districts and changes in the distribution are therefore easier to observe for the reader.

Methodology for reallocating funds to district level under portability

I aggregate the FY2015 district allocations under Title I Part A up to the state level, then divide state totals by the number of eligible children in each state. If all of Title I Part A were to be reallocated to vouchers (again, these vouchers would flow to public and private schools), this would yield the state’s voucher amount following each poor child to his or her school. I then assume that each school district receives that amount for each poor student enrolled in 2014-15: that is, I assume that no students take their vouchers to private schools. This simplifying assumption is also a conservative choice. It yields a lower bound, understating the potential impact of portability on federal funding for public schools, and showing how portability would matter even in areas with little private school presence.

District-level simulations for Maryland

Table 1 shows how portability would affect allocations across Maryland’s 24 school districts. Four would lose federal funding and 20 would gain funds. The four districts that would lose funds are: Baltimore City, Baltimore County, Montgomery County, and Prince George’s County. Baltimore City would lose an estimated $92 per eligible child, while the other districts losing funds would lose considerably less. Baltimore City currently receives 25 percent of the state’s Title I funds, even though it has slightly less, 23 percent, of the state’s eligible children. This is a function of how the formulas weight eligible children additionally in high poverty areas. Under portability, it would instead receive 23 percent of Maryland’s Title I dollars, matching its share of the state’s eligible children.

Table 1. How portability could affect district-level Title I allocations in Maryland

Public school district  Enrollment  % children
eligible for
Title I
Title I funds per eligible child, current rules  Title I funds per
eligible child,
portability state option 
Portability Title I per eligible minus current rules
Title I per eligible
Allegany County 8,865 20% $1,361 $1,668 $307
Anne Arundel County 79,518 8% $1,545 $1,668 $123
Baltimore City 84,976 31% $1,760 $1,668 -$92
Baltimore County 109,830 12% $1,675 $1,668 -$7
Calvert County 16,031 8% $1,169 $1,668 $500
Caroline County 5,592 22% $1,327 $1,668 $341
Carroll County 25,879 7% $1,198 $1,668 $470
Cecil County 15,681 13% $1,218 $1,668 $450
Charles County 26,258 10% $1,258 $1,668 $411
Dorchester County 4,796 27% $1,413 $1,668 $255
Frederick County 40,782 8% $1,288 $1,668 $380
Garrett County 3,858 21% $1,315 $1,668 $353
Harford County 37,536 9% $1,307 $1,668 $361
Howard County 53,685 6% $1,285 $1,668 $383
Kent County 2,106 20% $1,302 $1,668 $367
Montgomery County 154,434 9% $1,670 $1,668 -$1
Prince George’s County 127,567 13% $1,700 $1,668 -$32
Queen Anne’s County 7,724 9% $1,098 $1,668 $571
Somerset County 2,938 34% $1,525 $1,668 $143
St. Mary’s County 17,887 11% $1,203 $1,668 $466
Talbot County 4,630 15% $1,265 $1,668 $403
Washington County 22,327 15% $1,529 $1,668 $140
Wicomico County 14,545 22% $1,456 $1,668 $213
Worcester County 6,654 20% $1,333 $1,668 $336

Because the relationship between poverty rates and Title I grants is not linear in the formulas, the relationship between poverty rates and how a district would be affected by portability is not straightforward. In Baltimore City, about a third of children are eligible for Title I; the other three districts have 9 to 13 percent of children eligible. Like Baltimore City, Somerset County also has about a third of its children eligible, but is much smaller, with total enrollment just shy of 3,000. Somerset County would benefit from a shift to portability (gaining an estimated $143, 9 percent of its current allocation, per eligible student) while Baltimore City would lose about 5 percent of its current allocation per eligible student. The largest Maryland “winner” in terms of percentage increase in per eligible allocation would be the relatively small and non-poor Queen Anne’s County, whose allocation per eligible child would increase by about 50 percent.

Congress has been interested in the use of “number weighting” in the formula for Title I targeted grants, which allows weights by poverty counts or rates, whichever yield a more beneficial allocation to the district. Rural lawmakers dislike number weighting, and these results show why. Under portability and absent formula changes, a district with a high child poverty count would continue to generate higher Title I allocations per eligible child, but would be forced to share these gains with other districts across its state.

Districts currently have a lot of local control in dispersing Title I funds across their schools

School districts distribute Title I funds to specific schools they identify as “Title I schools,” subject to rules about how they may rank and serve their schools. Under these rules, districts rank their schools by need (they are required to use a percentage rather than count of economic disadvantage, and often use free- and reduced-price-lunch eligibility rates for this purpose) and then have discretion about how they wish to concentrate funds—that is, they choose how far down the list to go, subject to the requirement that they cannot “skip” a poorer school and then serve a wealthier school serving the same grade span. While some poor districts serve nearly all their schools, other districts choose to serve only one school. While child poverty and population data are used in the federal process of generating district-level allocations, districts may and often do use other data sources to determine school-level allocations, most notably free- and reduced-price lunch (FRPL) eligibility rates.

Districts report the Title I status of their schools as part of the National Center for Education Statistics’ Public School Universe (NCES PSU) Survey in the Common Core of Data. These data also report counts of students eligible for FRPL, and total enrollment by school.

Variation in school-level FRPL rates for Title I and non-Title I schools, Maryland

Table 2 uses these data to report measures of the distribution of FRPL eligibility rates across Title I and non-Title I elementary schools in each district in Maryland. I limit the analysis here to elementary schools because they are disproportionately likely to receive Title I funds compared to middle or high schools. The table reveals significant local heterogeneity not just in disadvantage, but also in how school districts exercise the discretion they have under the current ranking and serving rules. Title I schools are a diverse group, and their characteristics vary considerably both across and within districts in a state.

Table 2. The distribution of school-level free and reduced-price lunch eligibility rates in Title I and non-Title I elementary schools, by district (Maryland)

Public school district Title I elementary schools Non-Title I elementary schools
n 25th percentile 50th percentile 75th percentile n 25th percentile 50th percentile 75th percentile
Allegany County 8 60.3% 68.8% 72.1% 6 50.7% 53.6% 59.8%
Anne Arundel County 16 70.6% 78.1% 85.5% 65 14.4% 27.0% 40.5%
Baltimore City 116 88.6% 94.1% 96.2% 13 37.5% 53.4% 65.8%
Baltimore County 46 67.7% 73.1% 77.8% 62 14.7% 34.8% 47.6%
Calvert County 3 18.9% 36.4% 42.0% 9 16.8% 19.2% 23.4%
Caroline County 5 47.0% 50.3% 69.0% 0
Carroll County 4 39.4% 46.9% 55.4% 20 13.5% 20.7% 27.0%
Cecil County 7 56.7% 62.8% 67.0% 9 34.1% 38.3% 41.6%
Charles County 6 62.2% 64.3% 66.1% 15 26.6% 33.8% 41.9%
Dorchester County 4 71.2% 79.6% 86.4% 3 37.0% 40.7% 64.8%
Frederick County 9 42.3% 51.3% 72.7% 30 11.2% 15.6% 30.8%
Garrett County 5 59.5% 60.6% 69.1% 3 25.4% 35.7% 46.6%
Harford County 7 67.8% 76.5% 82.7% 26 13.0% 18.5% 33.1%
Howard County 12 41.3% 44.6% 55.8% 29 4.3% 9.3% 20.1%
Kent County 4 55.9% 60.8% 64.8% 1 48.9% 48.9% 48.9%
Montgomery County 28 70.5% 75.0% 81.4% 107 11.1% 24.2% 48.7%
Prince George’s County 64 82.0% 86.4% 90.3% 83 41.6% 55.9% 68.9%
Queen Anne’s County 3 34.4% 35.6% 59.9% 5 21.1% 23.2% 24.3%
Somerset County 3 94.4% 96.2% 97.6% 2 58.3% 65.5% 72.7%
St. Mary’s County 4 57.8% 61.9% 73.7% 14 22.7% 30.8% 34.6%
Talbot County 3 41.6% 54.1% 60.2% 2 28.8% 34.5% 40.2%
Washington County 8 69.0% 77.1% 92.3% 19 30.0% 38.7% 49.1%
Wicomico County 9 70.2% 83.9% 85.4% 7 44.4% 47.0% 51.6%
Worcester County 3 54.0% 59.5% 70.7% 2 32.8% 34.6% 36.3%

For example, in Baltimore City, the vast majority of schools get Title I funds—and are highly disadvantaged. Among the 13 elementary schools not participating in Title I, the median school had about half its students eligible for free- or reduced-price lunch; among the 116 Baltimore City Title I elementary schools, the median school had 94 percent of students eligible. Meanwhile, in Caroline, Frederick, or Talbot County Schools, the median Title I school had about the same free- or reduced-price lunch eligibility rate as the median non-Title I school in Baltimore City. How much any school stands to lose or gain from a switch to portability depends on how its district is choosing to divide program funds in the current regime, so this local heterogeneity matters.

Portability would change distribution of federal funds across public schools within districts

Under portability, the funds follow the student. Any school, public or private, would automatically become a Title I school once a single poor child enrolls.

We do not have public data on the amount of Title I funds at the school level now (though state reporting of such data is a new addition to ESSA). Simulating the impact of portability requires simulating district allocations of federal funds to the school level, both under the current regime and under portability.

Methodology for simulating school-level allocations under portability

To simulate school-level Title I funds under the current regime, I assume that districts allocate the same amount of Title I dollars per eligible student to each Title I school, regardless of grade span or FRPL-eligibility rate. This is not legally required, nor do I expect it to accurately describe the behavior of any given district; without school-level reporting of Title I funds, however, some sort of assumption is required. This assumption is not only simple and transparent, but also provides a conservative lower-bound estimate of how much redistribution across schools portability might entail.

Title I funds are currently allocated by the U.S. Department of Education to districts based on child poverty data, and frequently within districts based on (different) FRPL data. I calculate how school-level Title I funds would change on the basis of FRPL-eligibility rather than poverty based on available data. As I switch the metric from per-Title I eligible (i.e., from child poverty counts) as in the district-level calculations in Table 1 to per-FRPL-eligible student, the grant amounts shrink as more students participate in free and reduced-price lunch than are poor (and counted for district-level allocations).

School-level impacts of portability: simulations for Maryland

Under portability, each school would be affected both by reallocations of Title I funds across districts, as shown in Tables 1 and 2, and by reallocations across schools within its district. The reallocations across schools are generated by the old Title I schools sharing the district’s allocation with all schools in the district under portability, based on the number of eligible students per school.

Table 3 shows the percent of schools in each district reported to be participating in Title I in the 2014-15 NCES PSU data. Assuming that Title I funds are spread evenly per FRPL student across schools reporting Title I programs, the next column reports the mean Title I per FRPL amount in Title I schools (the simulated amount under the current regime). This ranges from $844 per FRPL pupil in a Title I school in Kent County, which serves 71 percent of its schools, to $2,917 per FRPL pupil in a Title I school in Carroll County, where 9 percent of schools are served with Title I funds.

Table 3. Simulation of school-level Title I allocations under current rules and portability, Maryland

Public school district Share of schools now receiving Title I Title I per FRL pupil in Title I school, current rules Title I per FRL student in all schools, portability  Change for current Title I schools in Title I per poor pupil from portability (portability minus current)
Allegany County 36% $1,362 $654 -$709
Anne Arundel County 13% $2,103 $475 -$1,628
Baltimore City 77% $874 $656 -$218
Baltimore County 29% $1,346 $495 -$851
Calvert County 12% $2,837 $639 -$2,198
Caroline County 50% $1,033 $688 -$345
Carroll County 9% $2,917 $649 -$2,269
Cecil County 25% $1,269 $578 -$691
Charles County 16% $1,867 $512 -$1,354
Dorchester County 31% $1,291 $710 -$582
Frederick County 13% $1,293 $523 -$770
Garrett County 42% $1,467 $883 -$584
Harford County 13% $1,840 $541 -$1,299
Howard County 16% $1,386 $516 -$871
Kent County 71% $844 $743 -$101
Montgomery County 14% $1,985 $480 -$1,506
Prince George’s County 36% $860 $376 -$483
Queen Anne’s County 29% $1,291 $654 -$637
Somerset County 33% $1,127 $684 -$442
St. Mary’s County 15% $1,586 $617 -$969
Talbot County 38% $1,162 $688 -$474
Washington County 17% $1,732 $582 -$1,150
Wicomico County 36% $1,410 $675 -$735
Worcester County 21% $2,416 $784 -$1,631

Under portability, each school would get the state-level allocation per FRPL-pupil multiplied by the number of FRPL students in the school. Table 3 then reports simulated allocations under portability, where the district-level allocations reported in Table 1 for Maryland are divided among FRPL students in all schools in each district, not just Title I schools. The final column shows how Title I funds per FRPL student would change for currently designated Title I schools in each district. This change is stronger in places where Title I funds are concentrated in a relatively smaller share of schools now. For example, in Baltimore City, where about three-quarters of schools already participate in Title I, portability would result in a loss of about $200 per poor student in schools now receiving funds. But in Montgomery County, where only 14 percent of schools get Title I now, those Title I schools would stand to lose about $1500 per poor pupil under portability.

For current Title I schools in Baltimore City and Montgomery County, decreases reflect both a reduction in the district’s total Title I allocation under portability (as in Table 1) and that the district’s funds would be spread more thinly under portability, across all its schools. Frederick County provides an interesting comparison. Like Montgomery County, its funds are now relatively concentrated, in about 13 percent of its schools. Unlike Montgomery County, portability would increase the size of the Title I pie it gets to divide among its schools (Table 1); its non-Title I schools are also less poor than in Montgomery County (Table 2). It is therefore unsurprising that Frederick County Title I schools would lose less, about an estimated $900 per poor pupil, compared to the $1500 per poor pupil in Montgomery County Title I schools under portability.

Policy implications

Lawmakers considering portability or other federal voucher programs must understand that the concept of federal dollars going into a “backpack of cash” that follows eligible students to the schools of their choice, whether public or private, is only part of the story. Portability would make significant changes to how federal dollars allocated at the state level flow across and within school districts, private schooling aside.

If portability returns to the policy discussion, it will likely be once again as a “state option” rather than a federal mandate. However, states may find themselves under considerable political pressure to adopt portability. This pressure would surely come from voucher advocates, but could also come from other groups who understand the nature of redistribution portability would entail. State policymakers who wish to switch over to portability should think carefully not only about reporting requirements and accountability for private schools under portability, but also about the details of the fiscal transition, such as hold harmless rates, that could allow high poverty public schools now served with Title I time to adjust.

The Every Student Succeeds Act declares the purpose of Title I to be: “to provide all children significant opportunity to receive a fair, equitable, and high-quality education, and to close educational achievement gaps.” Higher poverty schools and districts face greater challenges in meeting those goals: this is why Title I formulas now provide extra funds per poor student in poorer places. Under portability, this would no longer be true.

The author did not receive financial support from any firm or person with a financial or political interest in this article. She is currently not an officer, director, or board member of any organization with an interest in this article.

Author