6-7A Evaluate & Prioritize Expenditures to Enact the Instructional Priorities Pay for Priorities - 6 and 7: Identify Top Savings Options and Analyze Top Prerequisite Best Practices: None Key Points Savings Options SUMMARY Potential expenditures to enact the budget strategies described by a school district s Plan of Action must be evaluated both for: 1) the degree of alignment with the Plan of Action; and 2) the cost-effectiveness of the expenditure. A district should first inventory its current expenditures and any potential new expenditures, and then categorize the results by programmatic element. A programmatic element is a categorization of direct budgetary inputs that can be clearly associated with a service the school provides. After the inventory, a district can use the information to evaluate how proposed expenditures align with the Plan of Action. To evaluate an expenditure s effectiveness, a district can perform an academic return on investment (A-ROI), conduct a per-unit cost analysis, or analyze performance measures. A-ROI is the ideal form of costeffectiveness measurement, though it is not always practical. A-ROI seeks to quantify the total learning impact (learning increase multiplied by the number of students helped) per dollar spent. Per-unit costs (e.g., cost per student or teacher served) are a form of cost-effectiveness measure that is much easier to calculate. Per-unit cost can be derived for routine business and operational services as well as services that impact students directly. Performance measures indicate how much was accomplished, how well a service was executed, and the impact of that service on recipients. Related Award Program Criteria Criterion 6.A.1: Evaluating Spending for Alignment with Strategy (Mandatory). The Applicant has submitted documentation with its Supplementary Materials that shows the results of its evaluation of its programmatic elements for alignment with its Plan of Action. The Applicant can explain the process used and important conclusions reached in the Award Application. Criterion 6.A.2: Evaluating Spending for Cost-Effectiveness (Mandatory). The Applicant has submitted documentation with its Supplementary Materials that shows the results of its evaluation of its programmatic elements for cost-effectiveness. The Applicant can explain its approach to evaluating cost-effectiveness and any important conclusions reached in the Award Application. Criterion 6.A.3: Programmatic Element Inventory. The Applicant submits an inventory of its programmatic elements as part of its Supplementary Materials. Introduction An essential feature of a budget that best aligns resource use with student achievement is that spending on activities that support the district s Instructional Priorities is prioritized above spending on other items. 1 This necessarily requires that districts abandon an incremental approach to budgeting wherein the next year s budget is the same as last year s budget with changes around the margin to the degree necessary to distribute incremental revenue gains or losses among the district s subunits. It also implies that the district will take steps to actively determine the extent to which spending proposals align with the Instructional Priorities, and how cost-effective a given expenditure proposal is compared to other options for enacting the Instructional Priorities. This Best Practice document describes: 1
I. Getting ready to evaluate and prioritize the different options for spending available to the district by first understanding the options that are on the table. II. Finding opportunities to obtain the resources to pay for the new spending necessary to implement the Instructional Priorities. III. Weighing trade-offs between the benefits available from pursuing the Instructional Priorities against the cost associated with raising the resources to pay for the Priorities. IV. Overcoming constraints on changing how resources are used. V. Additional information and considerations when using the most advanced cost-effectiveness measures, academic return on investment (see Appendix 1), and discontinuing programs and services (Appendix 2). I. Getting Ready to Evaluate and Prioritize Background. As a prerequisite to evaluating proposed expenditures for alignment with Instructional Priorities, a district must understand, at a minimum, the cost of the activities needed to implement the Instructional Priorities. Ideally, however, the district will understand the full breadth of the programs and services it offers by conducting an inventory of its current expenditures. The GFOA recommends organizing this inventory by programmatic elements. A programmatic element is a categorization of budgetary inputs (e.g., personnel, dollars) that can be clearly associated with a service provided by a school or district. 2 Programmatic elements provide more insight into the services that the district and its schools are providing to students compared to raw object-of-expenditures (e.g., personnel, contractual services, materials, etc.). With this information in hand, the district can evaluate how well its inventory of current programmatic elements aligns with Instructional Priorities and how it compares to other spending options. It is likely that a district will find at least some of its programmatic elements do not align with the Instructional Priorities. These resources may be better spent elsewhere. Recommendation. The GFOA recommends that districts prepare to evaluate and prioritize their spending by developing an inventory of programmatic elements. Ideally, the inventory will include the purpose of the programmatic element, its cost, and some type of objective evidence of its effectiveness. The costs of each element should include all direct costs, including compensation of personnel and equipment they use. For evidence of effectiveness, a district should consider the three following methodologies: Academic return on investment (A-ROI). A-ROI seeks to quantify the total learning impact (learning increase multiplied by the number of students helped) per dollar spent. A-ROI is an ideal form of costeffectiveness measurement, but is not practical to use in all situations. Appendix 1 provides more information on A-ROI, including its limitations. The following two types of cost-effectiveness measures could provide substitutes, if A-ROI calculations are not possible. Per-unit costs. The budgets for routine and operational services, as well as programmatic elements that impact students directly, can be broken down into per-unit costs (e.g., cost per student served, cost per teacher impacted). 3 Methods for calculating per-unit costs are discussed in greater detail in Best Practice in School Budgeting, 2-A Identify Financial Position. While the cost per student served or teachers impacted is an imprecise measure of effectiveness, per-unit costs still represent an improvement for decision-support over aggregate expenditure figures when A-ROI information is not available. Performance measures. A performance measurement is a quantified indicator of how much was accomplished (e.g., number of students served), how well a service was executed, and/or the extent to which anyone (e.g., students or other clientele) is better off for having received the service. 4 Measures of how much was accomplished are generally considered the least powerful, while measures of the extent to which anyone is better off are the most powerful (these are also known as outcome measures). To illustrate, a programmatic element intended to increase graduation rates among an at-risk population of students should be measured by the graduation rate for those students, or a programmatic element designed to improve attendance should be measured by the percent of students with good attendance. 5 Performance measures like these could be used as the basis for an A-ROI calculation, or if calculation of A-ROI is not possible the measures at least provide a complement to cost data. For example, a programmatic element where performance measures indicate a positive outcome for students could also be examined on a unitcost basis to arrive at an estimate of cost-effectiveness.
II. The GFOA recognizes that many districts may not be able to develop a comprehensive inventory of their existing programmatic elements. At a minimum, therefore, all districts should at least identify the costs associated with implementing their Instructional Priorities. It is likely that this implementation will entail at least some new costs above what the district is budgeting. It is also likely that existing resources will need to be repurposed to make this happen. Hence, a district should identify both the total direct costs of implementing an Instructional Priority as well as the new costs that the district would need to fund. In instances where a district intends to implement an Instructional Priority by repurposing existing resources, it should note in the cost analysis what purpose the resources are being shifted away from. Finding Resources to Pay for the Instructional Priorities Background. Instructional Priorities often entail new costs and a district must find a way to pay for them. There are three basic ways in which a district might find these resources: Revenues. Raising taxes is an unrealistic option for many districts, but districts should still consider how their revenue structure will support funding the Instructional Priorities. Sunset programs. The district could discontinue certain programs and redirect the funding to the Instructional Priorities. Find efficiencies. The district could find ways to perform existing programs for a lower cost and direct the savings to the Instructional Priorities. Recommendation. The GFOA recommends that districts consider all three options described above for paying for the Instructional Priorities. Revenues. All districts should start by conducting an analysis of their current revenue sources, including a three- to five-year forecast, in order to get a sense of the revenues that are available to fund the Instructional Priorities. Districts might also consider options for raising new revenues. The GFOA recommends that districts adopt revenue policies that describe the desired characteristics of new revenues and the proper role of one-time versus ongoing revenues to guide its search for new revenues. 6 Districts might also consider adopting a debt policy to guide its use of debt to fund capital facilities, which might help optimize the use of debt versus current revenues to fund capital purchases. Sunset programs. A district should actively seek to identify and discontinue activities that are not aligned with its Instructional Priorities or that are not cost-effective. An inventory of programmatic elements is the ideal way to systematically look for activities that could be eliminated. A-ROI is the best tool to determine the cost-effectiveness of an activity, which should the primary consideration for whether or not to discontinue an activity. However, in the absence of A-ROI data, Appendix 2 provides a strategic abandonment tool that districts may be able to use to help with the review and sunset process. 7 A district should ensure that the programmatic element review and sunset process is transparent, especially the rationale for sunsetting a programmatic element. It should also make sure that all costs associated with the programmatic element are eliminated and/or reallocated to another programmatic element. Accommodations should be made for employees to the greatest extent possible, including transfer to available positions elsewhere in the district. Districts should also identify a clear timeline for reviewing and, where necessary, sunsetting programs. This can improve stakeholder support. Find efficiencies. All districts should actively investigate possibilities for providing existing services more efficiently. In Spending Money Wisely, the authors detail the top 10 saving opportunities for school districts and provide ideas for over 50 more. 8 Districts should start by conducting a high-level screening of these ideas to quickly identify those that may have potential. They should then undertake a more detailed feasibility assessment to get a better sense of the financial benefit available, the impact on student achievement (if any), the political feasibility, and the certainty of the financial gain relative to the complexity of implementing the idea. A broad cross-section of the district s management should be involved in evaluating the ideas. III. Weighing Trade-Offs
Background. A district must ultimately weigh its Instructional Priorities against the opportunities it has found to raise resources does the benefit available from the Instructional Priorities justify the cost to the organization to raise taxes, discontinue other programs, or change the way some programs are provided? The district s budget principles are the starting point for weighing these trade-offs because they describe the spirit that will guide decision making. 9 The district s decision makers should review and re-affirm the policies and principles before weighing the trade-offs. Using the principles as a touchstone, a district should develop a decision-making process that suits its unique culture and organizational structure. Recommendation. The GFOA recommends that districts use a structured judgment approach for weighing trade-offs, wherein the options under consideration (e.g., cutting a program, providing a program in a new way, implementing an entirely new program) are rated against various criteria. Many different approaches to rating can work (e.g., forced ranking, weighted scorings), but any system should include the following features: Transparent criteria. Clearly define the criteria that will be used to conduct the evaluation so that stakeholders understand what the options will be weighed against. The primary criterion should be the potential impact on the district s goals and consistency with the Instructional Priorities. Other criteria might include: Scalability. Can the activity start small and be easily scaled up if it proves successful? This is especially important for new or unproven ideas. Reversibility. If the expected benefit doesn t occur, can the decision be easily changed, or are there significant sunk costs? Options that have large sunk or fixed costs may be less attractive. Certainty of benefits. How likely are the expected benefits to occur? A risky course of action might be a lower priority than a more certain one. Ability to implement. Does the district have the time, people, skills, and other resources necessary to successfully implement the change required? Transparent scoring. Those performing the evaluation should be required to provide a written or at least verbal justification of their evaluation scores. Requiring judges to provide justifications may increase the consistency of their scoring and help move the process away from an intuitive process and toward an analytical one. 10 Documentation makes the process more transparent and may also help increase the overall quality by making the participants more conscious of their decision-making approaches. 11 Use of Data. Very high scores for any option should require some form of evidence to support the score. IV. Overcoming Constraints on Change Background. Making trade-offs often suggests new patterns of spending, but certain forces favor retaining past patterns of spending and constrain change. Examples of such forces include: Categorical funding limitations. Categorical funding sources such as Title 1 are often seen as highly constrained in their use, thus precluding the possibility of re-investing these funds in different types of activities. Legal mandates. The district may be subject to certain legal mandates that are presumed to require specific spending patterns, such as class size maximum, maintenance of effort, target or school-wide use of Title 1 funds, special education services related to IEP, or 504 requirements. Organizational culture. The culture of the district may possess a natural resistance to change. The need to redo individual education plans (IEPs). IEPs for special education students are difficult to change. Contractual or practical limitations on personnel transfer and reassignment. Labor contracts or management issues may pose an obstacle to re-assigning personnel. Recommendation. The GFOA recommends that districts consciously recognize that it is common for districts to overestimate (sometimes drastically) the level of constraint they are subject to. 12 The district should commit to investing the time to figuring out how to overcome these barriers and assesses possible solutions, such as:
Determine whether the constraint is real. Some constraints are less constraining in reality than they are imagined to be. For example, a legal mandate may not require the exact type of expenditure a district is making or a labor contract may not actually have provisions that restrict a certain course of action. Seek waivers. The district may be able to obtain a waiver or exemption from a particular constraint. For example, when trying to use categorical funds differently, write a letter to the funder stating how the district wants to use the money and why the district believes it is within guidelines. Ask for a written response addressing correct procedure. Get legal advice. Ambiguity in regulations does not necessarily mean that something can t be done. Get legal advice to help understand how the district can work within laws and regulations while still developing the budget that does the most possible to improve student achievement. Organizational change management. The district can communicate with individuals who are resistant to change to understand and address their concerns. Endnotes 1 An Instructional Priority is an overall approach for overcoming the challenges the district faces and achieving its goal. See Best Practice in School Budgeting, 4C Research & Develop Potential Instructional Priorities. 2 A programmatic element is similar to what is commonly termed a program in public budgeting literature (i.e., a set of activities with a common goal ), but is intended to be a little less precise in its application because the GFOA did not wish to describe program budgeting as necessarily being a best practice for school districts given that the state-mandated charts of accounts and reporting requirements might make development of a true program budget very difficult for a school district. Instead, the GFOA recommends that school districts seek to go beyond analyzing and budgeting just by objects-of-expenditure and that school districts impart a programmatic perspective into their budget, even if the analysis does not fully comply with the literature s definition of a program. Note: the definition of a program is taken from Robert Bland and Irene Rubin, Budgeting: A Guide for Local Governments (Washington, D.C.: ICMA, 1997). 3 The concept of per-unit costs in education is taken from Marguerite Roza, Now is a Great Time to Consider the Per-Unit Cost of Everything in Education in Stretching the School Dollar, ed. Frederick M. Hess and Eric Osberg (Cambridge, MA: Harvard Education Press, 2011). 4 Typology of measures taken from Mark Friedman, Trying Hard is Not Good Enough: How to Produce Measurable Improvements for Customers and Communities (Santa Fe, NM: FPSI Publishing, 2005). 5 Ibid. 6 See Shayne Kavanagh, Financial Policies (Chicago: Government Finance Officers Association, 2012). 7 Strategic abandonment tool taken from Hess, who adapted it from Heliodoro Sanchez, Strategic Abandonment Tool (Houston: Center for Reform of School Systems, 2012), http://www.crss.org/tll_files/documents?strategic%20abandonment%20tool.pdf. 8 Nathan Levenson, Karla Baehr, James C. Smith, Claire Sullivan, Spending Money Wisely: Getting the Most from School District Budgets (Boston: District Management Council, 2014). The book is also available for free online. 9 Further information on this topic is available in Best Practice in School Budgeting, 1A Develop Principles and Policies to Guide the Budget Process. 10 Thomas R. Stewart discusses research in forecasting science that shows that requiring expert judges to justify their opinion led to greater consistency and toward a more analytic process. See Thomas R. Stewart, Improving Reliability of Judgmental Forecasts in Principles of Forecasting, ed. J. Scott Armstrong (Boston: Kluwer Academic Publishers, 2001). 11 Forecasting science research describes the value of documenting how forecasts are made when expert judgment is a primary input into the forecast (as opposed to a forecast based solely on a quantitative statistical model, for example). See for example Derek Bunn and George Wright, Interaction of Judgmental and Statistical Forecasting Methods: Issues and Analysis, Management Science, 37, no. 5 (May 1991): 501-518. 12 Frederick M. Hess, Cage-Busting Leadership (Cambridge, MA: Harvard Education Press, 2013).
Appendix 1 About Academic Return on Investment While academic return on investment (A-ROI) is an ideal form of cost-effectiveness measure, school districts do not commonly used it. School districts may need to migrate towards more rigorous cost-effectiveness measures such as A-ROI, however, as the state and federal government, grantors, and other entities in the enabling environment for school districts require stricter controls over spending. 1 This appendix provides: Definition of A-ROI Benefits of A-ROI Calculating A-ROI When to use and not to use A-ROI An example of A-ROI Guidance on detailed application of A-ROI Definition of Academic Return on Investment (A-ROI) A-ROI is a tool to emphasize cost-effectiveness in budget decisions and help officials make more informed choices between different potential uses of resources. The basic formula for A-ROI appears below. 2 A-ROI = ((Learning increase) x (Number of students helped)) Dollars spent Benefits of Academic Return on Investment The key benefit of A-ROI is that it provides a clear measure of whether or not services are having their intended impact for a reasonable cost. More fundamentally, A-ROI tells a district if what it thought were good ideas for using tax dollars are actually working ideas. Further, using A-ROI makes explicit the following types of principles and policies in the budget process: Not all money spent with the intent to help children learn is necessarily effective. All money spent should lead to positive outcomes. Spending $1,000 to help a student learn is better than spending $2,000 for similar gains in learning. While some strategies may deliver a learning increase, these strategies also may be too expensive relative to other options to deliver similar increases. Money that isn t being used cost-effectively should be re-directed to other purposes. Legal restrictions on funding may constrain the extent to which ineffective spending or activities can be curtailed, but A-ROI helps encourage decision makers to curtail them to the extent possible. Calculating Academic Return on Investment The most challenging part of the A-ROI calculation is the benefit produced by the service. Benefits can be estimated in at least the following ways. 3 Run controlled tests. Especially germane to newer activities, the activity can be implemented as a pilot or experiment where the students benefiting from the activity are compared with a control group to estimate the level of benefit available. Look for natural experiments. Natural differences between how students are served within the district can provide an opportunity to observe the potential benefits from expanding an activity from one part of the district to others. However, a district must take into account student demographic and other characteristics when observing a natural experiment to arrive at valid conclusions. Formative assessments. Students can be tested for progress (e.g., learning growth) after participating in the activity being evaluated.
Establish and monitor other salient measures. For example, an activity intended to prevent dropouts should have a noticeable impact on the dropout rate. Costs are in the denominator of A-ROI. Costs should be full costs which include, at a minimum, the wages and fringe benefit costs of the employees performing the activities. If there are significant direct costs for equipment, outside contractors, and so forth, those costs should also be included. When to Use and Not to Use Academic Return on Investment A-ROI is an ideal form of cost-effectiveness evaluation and can be applied to a variety of different activities from specific professional development efforts, to small schools, to remediation and intervention. However, it is not applicable in all situations. Here are some guidelines for applying A-ROI judiciously. Apply A-ROI strategically. A district should not attempt to calculate A-ROI for all or even most of its services. It is likely that the time, cost, and effort needed to calculate A-ROI on such a broad scale would outweigh the benefit to decision making. Rather, A-ROI should be applied to programs that are high profile, high cost, or otherwise of special strategic significance. Apply A-ROI going forward, not backward. Gathering the historical data to calculate A-ROI can be very challenging. Districts will likely find it more beneficial to design the systems necessary to capture the required information for A-ROI going forward and then analyze A-ROI information once it becomes available. While this will require the district to wait for solid A-ROI information up to a year after the data collection systems have been implemented, it will result in much more accurate and credible information and will conserve the district s limited analytical resources. Plan ahead to use A-ROI. A-ROI analysis must be planned six to nine months ahead of when resource allocation decisions are made. Ad hoc A-ROI studies performed in response to the need to make budget cutbacks will not work. Calculating A-ROI needs to be a separate analysis step that occurs before resource allocation decisions are made. Do not apply A-ROI to services without direct academic benefit. Some services that a district provides are important but do not have a direct linkage to student learning. Examples include support services like payroll, transportation, or food services. Rather than attempting to estimate an indirect benefit to learning for these services, a district should use unit costs and performance measures, particularly measures of service outcomes, to evaluate cost-effectiveness. Example of A-ROI The following highly simplified example of A-ROI is from the work of school researcher Nate Levenson for a general education reading intervention and a special education reading intervention in a hypothetical district. The implication of this A-ROI finding is that this hypothetical district should seek to shift resources to the general education reading intervention. Not only does this intervention have a more cost-beneficial impact, it may even prevent the need for some children to go into the less cost-effective special education reading intervention. General Education Reading Intervention Special Education Reading Intervention Increase in student learning 1.5 years growth / year 0.8 years growth / year Number of students helped 100 40 Dollars spent per year $2,500 $5,000 Number of years of spending 3 10 Total dollars spent per student $7,500 $50,000 Total dollars spent over time $750,000 $2,000,000 A-ROI (per $10,000 spent) 4.5 0.02 Endnotes 1 Based on GFOA s discussion with school district practitioners and researchers. 2 Academic ROI materials are taken from Nathan Levenson, Smarter Budgets, Smarter Schools (Cambridge, MA: Harvard Education Press, 2012). 3 Ibid.
Appendix 2 Example of a Strategic Abandonment Tool The following is a simple tool that can be used to help districts to consider when to discontinue a program. A program is given a score for each criteria (1, 2, or 3 points) and the total points indicate whether the program s degree of conformance with the criteria is unacceptable, questionable, or acceptable. Criteria Acceptable (3) Questionable (2) Unacceptable (1) 1. The program maintains a clear metric for measurement. 2. The program's outcome or service rendered is measured frequently and without undue bias. 3. The program supports teaching and learning. 4. The program's service cannot be replicated otherwise. 5. The program's costto-service ratio is defensible. 6. The program is operated by the best personnel. 7. The program is necessary for the successful functioning of the district. 8. The loss of the program would cause a problem with a key stakeholder group. rendered is defined, and a clear metric exists to measure the program on a frequent basis. rendered is measured in an impartial manner on a regular (weekly or monthly) basis or better. The program directly supports teaching and learning through enhancing the educational setting, and faculty and staff can directly identify the tie between the program and instruction. The program's service is specialized and must be provided by specially trained personnel to ensure effectiveness, efficiency, and safety to all the program serves. The program's total cost divided by those it serves is better than what is found in similar districts without compromising the service provided. The program is administered by personnel who are familiar with the program and who stay within timelines and budget the vast majority of the budget year. Should the program not operate, the district would feel an immediate impact and the service would have to begin immediately for the district to maintain successful operation. A significant stakeholder group depends on this program, and loss would create a loss of faith. rendered is defined, but no metric to measure the program is available. rendered is measured regularly (weekly or monthly), but bias cannot be eliminated from the evaluation. The tie between teaching and learning and the program is related to evaluation. However, faculty and staff are not aware of the program's direct impact on instruction. The program can be provided by alternative personnel. But training and specialized supervision are necessary for the service to be conducted in an efficient, effective, and safe manner. The program's total cost divided by those it serves is within the normal estimations of districts with similar programs. The program is administered by personnel who are familiar with the program, yet personnel struggle to meet timelines or stay within budget. Should the program not operate, the district would function at a lessthan-acceptable level, and the service would have to begin anew within a month of its discontinuance. A significant stakeholder group is interested in this program, but loss would not create a loss of faith. is rendered unclear and undefined. rendered is not measured regularly or no documentation exists to verify accountability. There is no close tie or very limited tie between the program and teaching and learning. The program's service can be provided by alternative personnel with little to minimal training within the scope of the workday or workweek. The program's total cost divided by those it serves is beyond the norm for similar programs in similar districts or industries. The program is administered by personnel who are unfamiliar with the program or unable to execute the program's intent within acceptable timelines and cost. Should the program not operate, the district would continue to function with minimal disruption within a semester or an academic year. No significant stakeholder group is invested in this program. Total = 19-24=Acceptable 15-18= Questionable <14 = Unacceptable Program Evaluator: Program Evaluated: Date: