The Quest for a Targeted and Effective Title I ESEA: Challenges in Designing and Implementing Fiscal Compliance Rules

METHODS AND SOURCES

We begin by reviewing the policy history of ESEA Title I, leading up to the current compliance requirements, with particular attention to schoolwide programs. For this discussion, we draw on the law itself; the U.S. Department of Education's body of nonregulatory policy guidance documents, written for an audience of state and local education agency federal program administrators; the compliance circular produced by the Office of Management and Budget (OMB) providing more detailed guidance for the auditors; and technical assistance documents from nongovernmental sources that translate these complex original sources. This description and interpretation of Title I's key fiscal rules is a necessary foundation for describing the impact of these rules on practice and for understanding why practitioners may view the rules differently from the way Congress does.2

We also describe the federal single-audit process, which has been widely criticized for generating inaccurate findings. The outcomes of this process are critical to administrators, because the Department of Education may choose to follow up on negative findings with a process that can lead to a mandate to return funds to the federal government. To learn how often this feared outcome was realized, we submitted Freedom of Information Act (FOIA) requests to the Department of Education for all program determination letters related to Title I Part A for the calendar years 2011 through 2013; the request yielded 112 such letters to state and local education agencies, which we read to determine whether repayment of funds was required.

We selected a sample of target districts in four states, chosen for variety in embrace of school reform and for regional diversity. We limited potential districts to those with at least eight schools—this is the 25th percentile in the number of schools per district nationally when weighting by free and reduced-price lunch eligible (FRPLE) enrollment—and with a district-wide FRPLE rate of at least 30 percent. These precise cut-offs are arbitrary but motivated by the goal of ruling out districts with little program involvement and expertise. We divided these remaining districts into four quadrants, above and below their state's median enrollment and FRPLE rate, and randomly selected two districts per quadrant. We referenced each district's website to identify its Title I or federal grants administrator and contacted that individual in July 2014 via e-mail at least twice and if we received no response, via phone at least once. We also attempted to include the largest district in each state. Our sample of interviewees is necessarily limited to people who were willing to talk with us, who may or may not be representative of Title I administrators more generally. We emphasized in all contacts that all conversation would be off the record and the identity of the district and state would be obscured in our writing. We ultimately conducted eight interviews in the summer of 2014. One of the four states had administrators from four districts respond, one state had two, and the remaining two states each had only one.

SCHOOLWIDE PROGRAMS IN TITLE I: HOW ARE THEIR RULES DIFFERENT?

It did not take long for observers to note the unintended consequences of the new fiscal regime. Brenda Turnbull and Marshall Smith each provide excellent accounts of how the supplement not supplant regulations were perceived as fragmenting programming and requiring excessive administrator savvy and time to comply (Turnbull 1981; Smith 1986). Smith summarized the impact of supplement not supplant:In their attempt to implement this requirement, the U.S. Office of Education and state departments of education… issued regulations, guidelines, and other non-regulatory guidance and provided technical assistance to LEAs to help them design delivery mechanisms which were legal. The dominant choice to create a ‘clean’ fiscal trail was to create, in effect, a separate system within the school. Their goals were to keep the Chapter 1 teachers as separate as possible from the core program of the school, deliver Chapter 1 services in separate settings, and have separate technical assistance and reporting lines. By and large they succeeded. (1986, II-82)

Congress responded to these concerns by introducing in 1978 and later expanding eligibility for a schoolwide program (SWP) option for Title I schools serving particularly high concentrations of poor children.10 The fiscal rules for SWPs are significantly more flexible, allowing schools to use funds for services and materials benefiting the entire school, though they do face some additional compliance requirements: they must produce a plan, and must self-evaluate (in very loose terms) annually. Schools that receive Title I funding but do not operate SWPs, either because they are not poor enough to qualify for an SWP or choose not to take up that option, are referred to as targeted assistance schools (TAS) and are governed by fiscal rules similar to those governing all Title I schools in the 1970s.

When the schoolwide program was first introduced as an option, it was attractive for very few districts: a school needed to have 75 percent of its enrollment eligible and federal funds were conditional on a dollar-for-dollar match. There was little take-up of the schoolwide program option in the decade following its introduction, and pull-outs remained a key part of Title I (Wong and Meyer 1998). Marshall Smith references the District Practices Study of 1983, in which 18 percent of surveyed districts reported choosing pull-out designs for educational reasons, and 73 percent reported choosing pull-outs for compliance purposes (1986). His paper articulating problems with supplement not supplant had a specific policy recommendation—to expand use of schoolwide programs—and in 1988, the match requirement was eliminated.

The threshold for schoolwide program eligibility continued to be lowered over time, and take-up continued to increase. Title I schools are now eligible for schoolwide programs if at least 40 percent of their students are eligible for free or reduced price meals, and take-up is perceived to be generally high but variable by state and not universal. National data on the use of schoolwide programs are surprisingly hard to come by. The Public School Universe of the Common Core of Data includes variables for public schools on whether they are eligible for Title I and eligible to use the schoolwide model, but does not provide data on whether the schools participate in Title I, and if so, whether they use the targeted assistance or schoolwide model (see appendix B; Keaton 2012). The eligibility data reveal that two-thirds of public schools nationally are eligible for Title I programs, and of those, nearly three-quarters are eligible to operate schoolwide programs (Keaton 2012). Our interviews suggest that in large districts with many high-poverty schools, districts are likely to take up the schoolwide model in most if not all eligible schools.

However, we know little about the extent to which schools actually took advantage of the additional flexibility allowed in schoolwide programs in a way that promoted a more integrated and, potentially, productive approach to the use of Title I funds. That the schools exhibited so little initial demand for SWPs (as well as current incomplete take-up) and evidence of ongoing confusion about Title I's fiscal rules in general (such as the 2003 GAO report) both point to the possibility that the problems of fragmentation of funding and instruction identified in the 1980s may persist.11

Recall that supplement not supplant is meant to ensure that funds are targeted to participating students, not just schools. So what does this regulation require in a schoolwide program, where funds can be used to benefit all students in the school? Program administrators and auditors could refer to one of several sources of information to answer this question (see table 1).

The A-133 supplement (appendix C), referenced by auditors, begins by detailing three presumptions of supplanting (Executive Office of the President 2014). Using Title I funds for an activity is presumed to be supplanting nonfederal funds if any of the three are true: the activity is required by state or local law; the activity was funded with nonfederal funds in the prior year; or the activity was funded for children not participating in Title I with nonfederal funds in the current year. Auditors test these presumptions on a cost-by-cost basis; that is, the district must be able to show that every single good or service purchased with program funds meets each of the three presumptions. Informal discussions with state and local Title I administrators suggest widespread familiarity with the substance of the three presumptions.

However, the administrators we spoke with did not mention what the A-133 supplement explains immediately following the three presumptions: they must apply to targeted assistance schools, but schools operating schoolwide programs do not need to use federal funds to provide supplemental services, and instead can show that the federal funds are supplemental to the state and local funds that would have been allocated to the school absent its Title I program.12 This is clear from a close read of the full supplement, but could easily be missed. To be clear, the schoolwide programs of today have no legal reason to demonstrate the separateness between Title I programming and the core functioning of the school bemoaned in the past. In fact, eliminating this need—and the fragmentation of the instructional program it promoted—was a key goal of schoolwide programs.

Although the language in the A-133 supplement is relatively clear (if one perseveres in reading beyond the three presumptions), the clarity and completeness of the explanations provided in the Department of Education's nonregulatory guidance is more mixed. The department's 2006 “Designing Schoolwide Programs,” a fifty-four-page document, provides extensive guidance on how to comprehensively assess a school's needs, set goals, and devise an appropriate evidence-based plan. It does not, however, speak to any concern a district might have about whether particular evidence-based strategies for school improvement would be interpreted as supplanting other resources, and the word supplant does not appear in the document a single time. Appendix B reproduces an excerpt of the department's 2008 nonregulatory guidance on “Title I Fiscal Issues,” which clearly translates the supplemental funds test consistent with the A-133; however, the clear text appears in a section under “General Fiscal Issues” rather than in the sections for “Supplement not Supplant” or “Schoolwide Programs” and could easily be missed by federal grants administrators attempting to determine permissible uses of funds. The Department of Education recently clarified this point in a July 2015 “Dear Colleague” letter: “A schoolwide program school does not need to demonstrate that Title I funds are used only for activities that supplement, and do not supplant, those the school would otherwise provide with non-Federal funds” (2015, 5).

The extent to which the guidance is both salient and correctly interpreted by its readers in state and local education agencies is an open question. But the demand for professional interpretations of the guidance provides at least some evidence that federal grants administrators do not find the guidance completely straightforward.13 The department provides only a few hypothetical cases in its nonregulatory guidance, and does not make public the program determination letters (PDLs) summarizing the resolution of specific cases in which the single-audit findings prompt a more intensive response from the department, or even how many such cases exist.14 The private sector has responded to the demand for a larger sample of real cases: LRP Publications’ Title1 Admin service obtains PDLs via Freedom of Information Act (FOIA) requests and makes them available to its customers via paid subscriptions.15 Although many single-audit findings do not prompt PDLs, any finding ultimately resulting in repayment of funds to the department or involving significant “corrective action” will generate a PDL.

A PDL letter typically discusses more than one audit finding per agency. In some cases, agencies received more than one PDL concerning different audit periods or findings during the time period covered by our request. Table 2 summarizes the distribution of letters (rather than findings or agencies).

Over the three years, the total of 112 cases generated enough ED interest to result in a determination. Thirty-nine states (or their education agencies), the District of Columbia, and Puerto Rico received at least one such a letter in this period, the department requiring repayment in just under 10 percent of the cases. Whereas most states—which are tasked with monitoring compliance of their districts—received at least one PDL, only eight local school districts received PDLs, a vanishingly small share of districts receiving Title I funds. In those few cases when local districts did receive PDLs, however, they were more likely to be required to repay federal funds.16 Although school districts were highly unlikely to be subject to formal federal intervention following their single audits, the high rate at which the Department of Education intervened regarding audits of their state agencies means that state education agency (SEA) administrators have a clear interest in passing on the importance of compliance to the local education agencies (LEAs) they oversee.

WHAT DISTRICTS DO WITH THEIR TITLE I FUNDS AND WHY

Have schools and districts taken advantage of the additional flexibility that the schoolwide designation allows? Do they use Title I to fund the core instructional program? Unfortunately, data on school-level budgets are largely unavailable, and we know surprisingly little about whether schoolwide programs have altered how schools spend Title I funds. The American Recovery and Reinvestment Act of 2009 (ARRA) required school districts receiving Title I funds to report school-level per-pupil expenditures from state and local sources, but these data have a number of problems and ultimately provide little insight into what schools buy with those funds (Heuer and Stullich 2011).17

We are interested in what schools do differently when they have Title I funds, compared with what they would have done in the absence of the program. We are also interested in whether the answer to this question changed with the advent of schoolwide programs. We face at least two obstacles in addressing the question using the available data. First, finding an appropriate counterfactual is difficult, because Title I schools differ from non–Title I schools along a variety of dimensions. Second, we would need to observe how schools spend all their funds, but we only observe how Title I funds are spent in the application data. That is, the available data do not even allow us to say descriptively how spending in Title I and non–Title I schools differs.18 Nevertheless, we report what we can from the available data. Districts report how they use Title I and other federal funding in some detail when they apply for Title I funding. To gain some insight into how districts spend Title I funds in general and specifically in schoolwide programs we obtained and analyzed Title I application data for two districts. We also interviewed a small sample of district Title I administrators. We discuss our findings in turn.

For the reasons described, the application data do not allow us to determine how Title I funding affects what schools do. Nor is it straightforward to determine whether any particular Title I line item is funding the core instructional program; determining whether it is an efficient use of funds is of course even more difficult. But the data may provide some insight into a narrower question. Are Title I schools operating schoolwide programs taking advantage of the flexibility the law allows, spending money in ways they could not have thirty years ago—or in a targeted assistance program today? If schoolwide programs do not change Title I spending patterns, they may not have resolved the fragmentation problem they were designed to address. We find evidence that at least some districts do spend in schoolwide programs in ways that would not be allowed in targeted assistance schools. We also find some support for the idea that districts may use a “targeted assistance” approach to allocating and justifying Title I expenditures even when they operate schoolwide programs; this is consistent with “input from the field,” cited in the Department's 2015 “Dear Colleague” letter that there appear to be “some schoolwide program flexibilities that are not being used to their full extent.”

Both district A and district B are large, urban districts serving disadvantaged populations. Nearly all their Title I schools operate schoolwide programs. Consistent with other evidence about how schools spend Title I funds (Chambers et al. 2009), we find that both districts spent somewhat more than three-quarters of their Title I budget on personnel.19 To understand whether Title I purchases services that are part of the “core instructional program,” we need to know what the personnel purchased with Title I funds do. In district B, nearly half of spending for personnel (and over one-third of the total Title I budget) pays for regular classroom teachers for class size reduction. Class size reduction is not allowed in targeted assistance schools, so district B is clearly taking advantage of the flexibility the schoolwide program affords.20 District A, on the other hand, appears not to use any Title I funds to reduce class size (none of the descriptions include references to class size or smaller classes). District A does devote substantial Title I resources (about 5 percent of its budget) to extending the kindergarten day, a program that would not be permitted in a targeted assistance school.

Other line items are more difficult to classify. For example, the description for many line items indicates that teachers will provide “supplemental instruction” in core subjects, or “to support” core subjects; in many cases, the line item references “eligible students” or indicates that this will happen during the regular school day. Sometimes the description indicates that the role will be filled by a “highly qualified teacher.” These line items appear to refer to some form of small group or one-on-one supplemental instruction. This could be funding for traditional pull-outs, where students are taken out of class to work with a specialist or paraprofessional individually or in groups, or push-ins, where similar activity takes place inside the classroom. But we cannot tell from the description whether the supplemental instruction is well integrated with the regular instructional program or more separate, as in much-criticized pull-out programs. Similarly, both districts report substantial spending on instructional coaches, paraprofessionals, and interventionists—purchases that may seem more supplemental or separate, but if well-integrated with the regular instructional program could well be core.

Perhaps the most interesting difference between the two districts is that, in district A, line item descriptions commonly use language suggesting a targeted assistance approach to justifying spending. For example, a nontrivial share of line items mention “eligible students” or refer to “Title I students” or equipment, a distinction that is critical in targeted assistance schools, but not necessary in schoolwide programs.21 We want to emphasize (again) that we cannot make conclusions about the efficacy of any particular Title I line item or Title I spending overall. Rather, we view this exercise as pointing out the difficulty of understanding how schools use Title I funds based on what they report in the applications (though what they report there is voluminous), and showing that some districts clearly do take advantage of the flexibility schoolwide programs allow, but other districts and schools may not fully perceive or act on the added flexibility the schoolwide designation permits.

To better understand how Title I administrators decide how to allocate Title I funds, how they understand the rules that govern those allocations, and where they get information about those rules, we interviewed eight district Title I administrators. District administrators consistently referred to their state education agencies, or consultants retained by those agencies, as the key decision makers in judging which uses of funds were permissible. They understood that the principle of supplement not supplant is a federal one, not a state one, but perceive their states to be the arbiters of that concept.22 This points to the importance of targeting communication about what is allowable to state education agencies; we return to this point later.

Overall, the administrators were quite consistent in these interviews in characterizing their use of Title I funds as supplemental. We emphasize here and throughout that though some supplemental practices of the past (for example, pull-out instruction delivered by paraprofessionals who today would not be described as highly qualified) were widely criticized, we do not affix a value judgment to supplemental per se. For example, a district using Title I funds to provide struggling students with personalized support from a well-trained reading specialist may well be an effective use of funds. The extent to which it is effective depends not only on what the specialist is doing with the students, but also on how those activities align with the core instructional program the students experience in their regular classes.

We asked about common uses of funds in both schoolwide programs and targeted assistance schools, as relevant to district circumstances. Based on the activities they described, we conclude that their common use of the terms supplemental and intervention and avoidance of the term core does accurately reflect how they spend Title I funds, rather than being driven by reporting bias favoring activities they perceive to be compliant. The types of supplemental supports they described generally included expenditures on personnel: specialists or interventionists, instructional coaches, paraprofessionals, and in some cases, retired teachers working part time. When administrators reported using Title I funds for such personnel, we asked whether the personnel were working with students on content from the core curriculum, or on supplemental content: administrators typically responded that the content in the intervention, not simply its delivery format, was supplemental, though in some cases described the content as core support. These personnel provided services inside or outside the regular classroom during the regular school day (push-in or pull-out respectively) or before or after school via extended day programs. In some cases, students were served during the summer. Some administrators reported schoolwide programs using funds for regular classroom teachers for class size reduction. Professional development was also reported as a use of funds.

Administrators frequently referred to the interventions by their tiers in the response to intervention (RTI) framework.23 In this framework, tier 1 refers to what we think of as regular classroom instruction with ongoing screening to identify students in need of further supports, tier 2 is the first line of response, often in a small group, and tier 3 is the most intensive, individualized level intervention. RTI is a permissible and encouraged use of funds, and provides another example of how difficult it is to classify Title I expenditures as comprehensive versus fragmented, or core versus supplemental, in their contributions to school improvement. One can imagine a school implementing RTI faithfully as viewing all three tiers as essential parts of individualized, core instruction. It is also easy to imagine today's administrators describing programs much like the pull-out programs of the past—so distinct from the core curriculum as to be disjointed—as tier 2 or tier 3 interventions. In practice, the administrators we spoke to not only consistently referred to their Title I interventions as supplemental, but also mentioned their use of supplemental materials as separate from the regular curricular materials. One administrator clarified that these materials “are only allowed to be used by our intervention team,” as required by supplement not supplant in a targeted assistance school.

Administrators often described supplemental instruction via an extended day or extended year. In each case, we asked whether this was a uniform schedule for the entire school or an option for participating students (in the case of schoolwide programs, potentially for all students in a school). In all cases, the extra hours were optional, so regular classroom teachers could not expect all their students to receive any particular content via that program; extended day programs were often described as providing tutoring or supplemental educational services, but never core instruction.24 Again, seeing a line-item description on a budget (for example, “extended day ELA instruction”) does not communicate the full context needed to judge the likely efficacy of the funded activities or whether they represent an improvement over the old style of Title I services fragmenting educational programming. Some schoolwide programs offered extended day options for all students, and it is clearly legal to do so; targeted assistance schools are limited to offering this to their Title I–identified students. Yet when describing a schoolwide program offering before- and after-school tutoring to all its students, one administrator concluded, “We're probably breaking some rule.”

The interviews not only were helpful in decoding the types of expenditures described tersely in the application budgets, but also reveal why districts had chosen to allocate their funds as they did. Administrators rarely mentioned concern with future audits, but often mentioned the need for approval from their state education agency or its consultants during the application phase. One said, “They'll come back to me and say, listen, this is what you need to say, and then we'll fund it.” Similarly, they described their sources of information as coming from their SEAs, via webinars, PowerPoint presentations, and regional meetings of LEA Title I directors with state representation. They rarely referenced the federal Department of Education as a source of information about permissible uses of funds.

Administrators consistently described their programming choices as driven by the desire to improve student achievement for their struggling students. We asked administrators whether they had used Title I to support a number of specific activities. For activities they did not support, they often commented that their decisions were based on their expectations about what would improve student achievement. For example, the sampled districts did not use Title I funds to support arts instruction, nor did they have any interest in doing so. They also reported little to no use of funds for class-size reduction, one of the most intuitively obvious options opened up by the schoolwide program designation. Several described class-size reduction as not supported by research as an effective way of increasing achievement;25 others noted that it would take funds away from existing Title I programs.26

Some administrators described these choices as constrained by concerns about supplanting. For example, one administrator explained that she does not use Title I for software because it would be “hard to buy a computer program for a school with Title I one year and then expand to the rest of the district the next year.” Those using Title I funds in high schools reported that they did not use funds to support initial enrollment in credit-bearing courses to avoid supplanting, though some used funds to support credit recovery programs. Another noted more generally, “It's hard because we can't add to current programs, or enhance them, because it might not necessarily be viewed as a supplement.” Multiple administrators said concerns about supplanting had deterred them from funding school nurses or other health initiatives they would have liked to pursue.

District policies also constrain them: several administrators reported that they would not consider using Title I to upgrade a core curriculum in a school (such as reading) because then the school would not be using the same curriculum as the rest of the district. They also reported that they were encouraged by their state agencies to buy “people” rather than “things” with Title I, and that technology in particular was “very hard to keep track of.” One administrator described how in past experience as a school principal, “We were always encouraged to spend it on people for support, and the rationale that was given to us was that that's how the state prefers it, it's easier to get approved that way.”

We asked administrators whether federal or state rules prohibited them from doing anything they would otherwise like to do with their Title I money. The answer was almost uniformly no—that is, the administrators did not name any specific activity that fiscal regulations had prevented them from doing.27 Within this uniformity, however, were two clear camps: those who used their discretion to move around funds creatively within the limits of the law, and those who identified strongly with Title I's compensatory mission and felt that using funds for other (nonsupplemental) activities would dilute the program as experienced by disadvantaged students. Those in the first camp often described themselves as creative and those in the second emphasized transparency and fidelity to the program. Federal and state rules and regulations may both constrain and enable local implementers. Variation in how local implementers interpret and exercise discretion has been documented and studied in many settings (Hill and Lynn 2015). Aside from the standard principal-agent concerns, in this setting we argue that the goals of the (federal) principal historically have been poorly communicated to the (state and local) agent.28 It will be interesting to observe how very recent clarifications at the federal level affect state and local behavior going forward.

Of the eight districts represented in the administrator interviews, three districts used schoolwide programs in all of their Title I schools, and one large district did so in nearly all of them. Schools in one district operated only targeted assistance programs, despite eligibility for schoolwide status. This interview provided some interesting insights. The Title I director would like the schools to move to schoolwide programs, but resistance due to job security concerns is evident among current Title I staff at the schools. The director also expressed mixed feelings about whether switching to schoolwide programs would benefit the schools: “Tier 1 instruction is the problem, so a schoolwide program might not fix it.” This statement suggests that the director may not be aware that Title I can be used to support tier 1 (applied to all students) instruction in a schoolwide context.

In the remaining three districts, schools operated a mix of schoolwide and targeted assistance programs. Conversations with these district grants managers proved particularly useful in contrasting the activities supported with Title I in the two types of programs within a district.29 Schoolwide programs consistently did the same things as targeted assistance schools but offered some additional services. Key differences between schoolwide and targeted assistance use of funds in these three anonymized districts follow.

1. District X described using Title I in all its program schools primarily to provide extra academic support during the school day, via both push-in and pull-out modes. In schoolwide programs, this support comes from certified teacher interventionists. In targeted assistance schools, it is provided by highly trained paraprofessionals.

2. District Y reported using funds similarly in each of its Title I schools, for instructional support specialists, including a certified teacher who oversees a team of highly qualified paraprofessionals, and for an instructional coach. This model is followed by both schoolwide and targeted assistance programs; both types of programs identify individual students to be served by Title I using state and district assessments. Additional expenditures in schoolwide programs not found in targeted assistance schools included Chromebooks, a social worker for the school, and tutoring (by teachers) before and after school.

3. District Z has one SWP and one TAS. In both schools, Title I funds “mostly teachers”—math and literacy supplemental instructors providing Tier 2 (typically small group) interventions. They describe the teachers as serving students who need a “short-term boost of academic support … who function just below proficiency.” In the SWP, the specific students served change throughout the year as individual needs evolve, but in the TAS program they “must designate the kids” and do “not have fluidity to move kids in and out of support throughout the year.”

Overall, administrators across the three districts attributed most of the additional services in schoolwide programs to their higher funding levels stemming from their eligibility ranking within their districts, rather than to flexibility afforded by the schoolwide model. One administrator noted the flexibility of schoolwide programs has made it easier for them to purchase certain supplies compared to their targeted assistance counterparts, including online licensing for remediation programs.

CONCLUSIONS: FISCAL RULES AND EFFECTIVE SPENDING GOING FORWARD

ESEA is reaching its half-century mark in contentious times. The law is widely recognized as essentially inoperable as a funding mechanism, prompting widespread use of the waiver process under ESEA Flexibility because Congress has been unable to reauthorize a more functional version of the law. The political obstacles to reauthorization are primarily related to the strings attached to Title I funds, particularly resistance to state accountability systems. By comparison, the issues related to Title I's fiscal rules discussed here have received little attention. Yet however the debates about testing and standards are resolved, the questions we raise about how to optimally monitor Title I spending will remain critical. They also apply more broadly to federal education policy, as other programs have their own fiscal rules.30

The problem of targeting funds while promoting quality is widely recognized as a thorny one.31 All parties involved clearly expend a great deal of effort trying to get things right. The federal government tries to develop and enforce appropriate compliance regimes, introducing potential sources of local flexibility via policies such as schoolwide programs and the ability to consolidate federal funds. States and districts try to best serve their students while adhering to those complex and evolving sets of rules, devoting considerable staff time to documenting their compliance and formally associating particular expenditures with permissible titles (“moving money around”). In our interviews with district Title I administrators, several of the more creative told us of their efforts linking costs to pots of money they perceived to be legal and away from Title I, where they feared they would be supplanting; in multiple cases, in fact, the specific expenditures would have been permissible uses of Title I funds.32 Similarly, more cautious Title I directors often described roads not taken because of concern about supplanting. In both situations, misperceptions about the rules were costly to the district, either in terms of administrative burden or by preventing a perhaps more effective allocation of funds.

States also have developed their own categorical programs, often with compliance requirements similar to the federal ones. This creates a still more complex problem for district administrators needing to show that multiple funding streams each distinctly contribute to educational inputs. At this point, even if districts moved away from justifying the supplemental nature of Title I expenditures in schoolwide programs on a cost-by-cost basis, their discretion over their budgets would still be significantly restricted by the amassed compliance requirements over a variety of federal and state categorical programs (see, for example, Roza 2010; Hanna 2014).

We concur with Duncan and Murnane when they write, “The challenge is to devise organizational structures that provide high-poverty schools with the resources, knowledge, and freedom to choose the collection of supports they need” (2014, 136). To the extent that Title I's fiscal rules—and those of the web of other federal and state categorical programs contributing to district budgets—impinge on that freedom, we have two concrete recommendations. First, replace existing guidance with new, clear, and concise guidance and disseminate it. The dissemination challenge is significant, requiring a concerted effort by the federal government to get word to state education agencies who in turn must get the word out to local districts. Guidance should reach not only Title I personnel in districts and schools, but also others involved in allocating instructional resources. Clear guidance would empower districts to push back against state agencies that require them to remove legally permissible uses of Title I funds from their federal funding applications, and to question their preliminary single-audit findings before the results are finalized and passed on to federal authorities. It would also empower school district staff more closely associated with the work of instruction itself—in offices of teaching and learning, or curriculum and instruction, for example—to question district-level Title I administrators when they deny them access to funds on the basis of supplanting. Finally, it is critical for this guidance to reach auditors. Decision-makers need to know what they are allowed to do, and that legal uses of funds will not yield inaccurate and damaging audit results.

Our second recommendation relates to the challenge districts face in handling multiple streams of state and federal categorical funding, each with their own compliance requirements. Title I schoolwide programs are permitted to consolidate their Title I funds with other federal, state, and local funds so that they do not have to maintain distinct fiscal trails for each separate program. But a recent survey found only 6 percent of districts operating schoolwide programs took up this option, and found that “state or district accounting rules and fear of potential audit exceptions were major barriers to consolidation of funding” (Chambers et al. 2009, xxvii). Consolidating funding is not the explicit focus in this article, and our recommendation here is correspondingly broad: states and local governments should reconsider the impact of their rules on schools’ ability to use funds productively. The U.S. Department of Education requires each state education agency to “eliminate State fiscal and accounting barriers so that these funds can be more easily consolidated” (2015, 8). The federal government and national nongovernmental organizations could potentially play an important role in providing technical assistance in these efforts. We view both these policy recommendations as necessary but not sufficient for addressing the big problem—the lack of instructional capacity in school districts. We do not offer a magic bullet for building this elusive capacity, but instead hope that reducing the complexity of compliance and enhancing flexibility will contribute to creating an environment where efforts to do so have a better chance of flourishing.