This interim report to the Joint Subcommittee on Higher Education Funding Policies summarizes the work MGT of America, Inc. has completed to date in conjunction with the staffs of the Senate Finance Committee and House Appropriations Committee since meeting with the Subcommittee in October, 1999. This report includes the following:

• an overview of the materials presented by MGT to the Subcommittee in October, 1999;
• a summary of work activities completed since October, 1999; and
• the current status of the development of initial base funding guidelines for instruction, academic support, institutional support, student services, and operation and maintenance of plant and preliminary recommendations.

Because this is an interim report, the information presented and current status of activities should be viewed as preliminary and subject to change as the work product becomes more refined over the next several weeks.

In October, MGT presented a paper to the Subcommittee entitled A Framework for Virginia Higher Education Base Funding Guidelines: Concepts, Structural Issues, and Alternatives. Major topics of the paper included the following:

• desired characteristics of base funding guidelines;
• an overview of guideline usage, design, and trends among the states;
• an overview of the program areas to be covered by Virginia’s base funding guidelines; and
• the initial proposed framework for the base funding guidelines.

Each of these topics is briefly highlighted below.

2.1 Desired Characteristics of Base Funding Guidelines

Over time, a number of researchers in the area of higher education finance have offered their concepts regarding desired characteristics in state higher education funding formulas. Frequently, what is offered as the "desired characteristic" is in direct response to a perceived shortcoming of a particular state’s funding model.

Fourteen characteristics, listed and summarized in Exhibit 1 in no particular order of importance, often tend to be in opposition to one another. For instance, the desire to have a simple-to-understand funding formula may preclude features that might contribute to a greater degree of equity (e.g., more detailed sub-categories to reflect institutional differences). Similarly, a formula that is responsive to changes in enrollment levels may not be able at the same time to provide the desired level of stability. In keeping with these characteristics, one of the Subcommittee’s goals through this process is to attempt to establish an equitable funding model for all of Virginia’s public colleges and universities, while recognizing the unique aspects of the institutions.

2.2 Trends in Funding Guideline Usage and Design

MGT’s review of current higher education funding guideline usage and design across the states indicates the following:

• more than one-half of the states (27) currently use funding guidelines or formulas;
• states may use funding guidelines for only one or several different program areas; and,
• while there are broad similarities in guideline design among the states, each state’s methodology reflects its own specific situation or needs.

This last point is especially relevant for the Subcommittee’s work as it considers the development of base funding guidelines for Virginia’s public institutions of higher education. States may borrow basic guideline design features from other states, however in the end, there is no one best funding guideline methodology. Rather, it is more critical that a state’s guidelines reflect its own context and funding policy goals.

Some emerging trends in funding guideline design and usage among the states include the following:

• a search for more simplified and streamlined approaches to funding models
• greater use of non-guideline funding categories; and
• an increasing focus on quality and performance.

The second trend indicated (i.e., greater use of non-guideline funding categories) reflects a realization that, increasingly, there are unique programs and state policy priorities that cannot be adequately funded solely through the use of mathematical guidelines.

Five educational and general (E&G) program areas are being considered in the base funding guideline development process. They are as follows:

• Instruction
• Academic Support
• Institutional Support
• Student Services
• Operation and Maintenance of Plant

Exhibit 2 below provides examples of the types of activities and other expenditures covered under each of these five programs.

In developing funding guidelines, it is important to first isolate those factors that serve as "cost drivers", and therefore funding requirement drivers, for each program area. Some of the major "cost drivers" that have been recognized by other states in developing funding guidelines are shown in Exhibit 3 for each of the five program areas.

2.4 Initial Proposed Framework for Base Funding Guidelines

The process of designing a set of higher education funding guidelines is not unlike the process an architect goes through in designing a building for a client. There are some basic structural issues common to all funding guidelines (and buildings), which can be taken as a given by the designer. For example, all building designs need to consider the basic laws of physics to ensure structural soundness for the facility. Likewise, the design of funding guidelines needs to consider both the technical (e.g., data systems availability) and practical (e.g., political) considerations (and limitations) within the state to ensure that the guidelines are ultimately workable.

Beyond that, however, there is a gray area where the designer and client must work through an iterative process in order to "flesh out" the details. As indicated earlier, there is no one "right" funding guideline methodology just as there is no one "right" building design. As such, the designer must first consult with the client to determine the desired outcome before putting an initial proposal together for review. From that point on, the proposed structure is refined until it becomes a completed product in the eyes of the client and designer.

Exhibit 4 presents the initial framework for Virginia’s higher education base funding guidelines that we proposed in October. This framework is based on our discussions with legislative staff as well as our own experience. The purpose for such a framework is to provide that starting point for the funding guidelines from which to "flesh out" and refine. There are three underlying assumptions inherent in this framework:

1. The new base funding guidelines should complement, and not replace the Commonwealth’s existing funding policies for higher education (e.g., the faculty salary benchmark process). The Commonwealth has invested considerable time and effort in developing and refining its existing higher education funding policies over the past several years. Thus, any new policy initiative should build upon and not eliminate the prior accomplishments of the state.
2. To the extent possible, the guideline factors would be developed through an assessment of actual experience or "best practices" nationally. Given that Virginia’s colleges and universities compete in a national marketplace for students, faculty, and staff, institutional funding factors should reflect "industry standards" nationally as well.
3. Not all institutional resource requirements will, or should, be met through these base funding guidelines. For example, unique institutional programs (e.g., VIMS at the College of William and Mary, Agricultural Extension at Virginia Tech and Virginia State University), first professional medical education, hospitals, and other areas with special funding needs would continue to be funded outside of the base funding guidelines.

Further discussions with legislative staff, as well as with Subcommittee members at the October meeting have reinforced the importance of these assumptions.

Since the meeting with the Subcommittee in October, MGT has conducted various research activities in reference to the initial proposed framework for Virginia’s base funding guidelines that was described previously. Specifically, we have:

• gathered information on student/faculty ratios by discipline and level used by other states in their funding formulas or generated as the result of a special data collection;
• conducted several statistical analyses of national data on the relationship between institutional expenditures in academic support, institutional support, and student services and related cost drivers for each area (e.g., enrollment levels, academic program array, etc.);
• developed a data request instrument for Virginia institutions in conjunction with legislative staff in order to gather supplemental information;
• met with a technical advisory group comprised of institutional representatives in November, 1999 and January, 2000 in order to review and discuss the results of our work to date; and
• continued to refine our analyses and guideline framework.

The next section of this report presents more detailed descriptions of our work on the guidelines for each of the program areas.

This section of the paper presents an overview of the work to date, current status of the base funding guideline development, and preliminary recommendations for each of the five program areas:

• Instruction
• Academic Support
• Institutional Support
• Student Services
• Operation and Maintenance of Plant

Because the methodology for addressing the development of guidelines for academic support, institutional support, and student services has been similar, these three program areas will be discussed together as "support services".

4.1 Instruction

As indicated earlier in section 2.3 of this report, there are two sub-areas within the instruction program for which base funding guidelines are to be developed: instructional faculty salaries and support staff salaries/non-personal services expenses. Both are discussed below.

Instructional Faculty Salaries. As indicated earlier, the proposed methodology for instructional faculty salary base funding guidelines for each Virginia institution is as follows:

In other words, this approach results in an instructional faculty FTE staff number for each Virginia institution that can be multiplied by the institution’s specific faculty salary rate to arrive at an instructional faculty salary base for the institution.

From the beginning, a guiding principle for this approach is that it would recognize differences in instructional staffing needs among the different academic disciplines and levels of instruction (e.g., lower division, upper division, masters, doctoral, first professional). As such, we looked for potential taxonomies of academic disciplines by level used by other states that would help to inform the design process. The initial models included for analysis included both states where pre-determined student/faculty ratios for different disciplines and levels were used in funding formulas (e.g., Connecticut) as well as states that regularly collected data on actual student/faculty ratios by discipline and level (e.g., Wisconsin). In total, there were almost 50 separate discipline areas represented in the analysis. The results are included in Appendix A.

We also reviewed the related accreditation standards on instructional staffing requirements set forth by several programmatic accreditation organizations, representing the following areas:

• Allied Health
• Architecture
• Business
• Engineering
• Law
• Teacher Education

The purpose of this review was to determine whether there were minimum student/faculty ratios that were viewed as "industry standards" for certain fields of study (particularly the professional programs). The relevant accreditation standards of those organizations are shown in Appendix B. In general, while some allied health fields have specific recommended ratios, most accrediting organizations seem to afford institutions a high degree of discretion in staffing their programs, provided that academic quality is maintained.

The materials in Appendix A were reviewed with the technical advisory group at the November meeting. The initial consensus of the group at this meeting was that the instructional staffing guidelines for Virginia should be based on a more aggregated grouping of disciplines (i.e., fewer) to reduce complexity in the usage and application of the guidelines. However, the group also desired to maintain discrete levels of instruction in the guidelines (i.e., lower division, upper division, master’s, doctoral, and first professional-law).

Further research found two examples of more aggregated discipline groupings in use in Georgia and Wisconsin. The Georgia discipline groupings are shown in Exhibit 5 and the Wisconsin discipline groupings are shown in Exhibit 6. Both have been crosswalked to the discipline groupings used by Virginia institutions in reporting data to SCHEV. As indicated, the Georgia taxonomy has three discipline groupings (plus one for remedial programs), while the Wisconsin taxonomy has six discipline groupings. Further, while not indicated in this exhibit, the Wisconsin taxonomy recognizes the full range of instructional levels (i.e., lower division, upper division, master’s, doctoral, and first professional), while the Georgia approach has only three instructional levels (lower division, upper division, and graduate).

Both of these options were reviewed and discussed with the technical advisory group at the January meeting (see Appendices A-5 and A-6). There was general consensus on using the University of Wisconsin System’s six broad discipline groupings as the basis for the student/faculty ratio base funding guidelines, and in using the related UW System student/faculty ratio data if the data are appropriate, based on further investigation. The group was also interested in investigating further the appropriateness of having a separate discipline grouping for business, and the community colleges were interested in having a separate grouping for occupational technologies.

Preliminary Recommendation. At this point, MGT recommends further investigation into the application of the Wisconsin taxonomy to the Virginia base funding guidelines, with possible augmentation as suggested by the group. The schematic for the student/faculty ratios is shown in Exhibit 7 below. As indicated, the Wisconsin taxonomy meets the desired criterion of having a more aggregated grouping of disciplines while retaining differentiation by instructional level. Further, as mentioned earlier, the University of Wisconsin System collects and publishes actual data on student/faculty ratios according to this taxonomy that could be used to inform the development of the Virginia guidelines if found to be appropriate.

Non-Faculty Instructional Costs. As discussed earlier, the initial proposal to fund instructional faculty salaries within the evolving concept for the development of new base funding guidelines for Virginia’s public colleges and universities is through a combination of student/faculty ratios and peer faculty salary rates. However, the process for funding instructional support staff salaries and instructional non-personal services expenses (referred to as "non-faculty instructional costs" in this paper) has not yet been determined. This is a particular issue of concern to Virginia’s public colleges and universities who feel that these costs have been underfunded in recent years. There is also a belief among these institutions that these costs will continue to grow in the future, driven primarily by a rapidly increasing use of technology in instructional delivery.

There are at least six different methods that could be used as guidelines for funding non-faculty instructional costs (none of which are necessarily mutually exclusive):

1. An add-on percentage to total instructional faculty salary requirements generated for the institution (i.e., X% of faculty salary base).
2. Staffing ratios driven by the number of instructional faculty generated through the student/faculty ratios (e.g., 1 support staff FTE for every 4 faculty FTE).
3. A dollars per FTE student approach.
4. A dollars per student credit hour approach.
5. A "base-plus" approach which adjusts the previous year’s non-faculty instructional cost base by an inflation and/or enrollment growth factor.
6. A "zero-based" approach for non-faculty instructional cost components.

Further, from a conceptual standpoint, Methods #1 through #4 could potentially be differentiated by discipline area and/or institutional type, and or include a recognition of fixed costs/economies of scale.

Each method has its own relative advantages and disadvantages. Method #1 is quite straightforward, although it suffers from having no direct linkage to instructional workload at an institution. Among Methods #2 - #4 however, the primary advantage is a direct link to instructional workload. However, these methods suffer from a lack of existing actual data with which to calibrate the guidelines.

While Methods #5 and #6 are not "guidelines" in the strictest sense of the term, they are alternative budgeting methodologies that the General Assembly could choose to adopt for funding non-faculty instructional costs. There are significant disadvantages inherent in each approach, however. The primary disadvantage in Method #5 is that it may serve to perpetuate past funding inequities and inadequacies among and within institutions. The primary disadvantage to Method #6 is that any "zero-based" budget approach is very time consuming to implement and maintain.

An equally important issue to be resolved in establishing base funding guidelines for non-faculty instructional costs is the data to be used in calibrating the guideline. Because there is no central source of data nationally, there are essentially three data sources available in establishing the guideline:

1. Obtain relevant historical data from Virginia’s public colleges and universities.
2. Borrow/adapt guideline factors used by other states
3. Conduct a special survey of institutions and/or systems in other states to collect relevant actual data (or survey current peers)

As with the guideline methods discussed in the previous section, these three data sources are not mutually exclusive (i.e., different data sources could potentially be used for different cost components).

The primary advantage of the first two data sources is their availability. Their disadvantages are that they may serve to perpetuate funding inadequacies and inequities (#1), and may not be relevant to the unique needs of Virginia’s public colleges and universities (#2). The primary advantage of the third potential data source is the ability to tailor the data to a desired guideline method, as opposed to "fitting" the guideline method to available data. Its primary disadvantages are the time involved in data collection (significant), and the ultimate risk of a low response rate from those institutions and systems that are surveyed.

Preliminary Recommendation. After considering the various pros and cons of each alternative, MGT recommended that a special survey be conducted to determine current funding patterns in other states relative to non-faculty instructional costs. The survey was developed and sent to selected public university and community college systems nationally in February – March, 2000, and collected comparative data on non-faculty instructional support staffing and costs. The systems included institutions that are current peers of Virginia’s four-year institutions and community colleges. Responses were received from nine systems, representing 174 institutions. MGT is currently analyzing the data provided in these responses.

4.2 Support Services

As indicated earlier, to date MGT has used a similar methodology to research the development of base funding guidelines for academic support, institutional support, and student services. The general methodology used has been a statistical analysis of the relationship between unrestricted institutional costs in each of the three program areas and potential "cost drivers" through linear regression modeling. The cost drivers explored for the three program areas through these models are shown in Exhibit 8. The focus has been on unrestricted expenditures in order to exclude those costs that are generally outside of the institution’s discretion (e.g., grant-funded activities). The ultimate goal of these analyses is to generate funding factors that have been shown to have a statistical relationship with each of the three program areas. Further, in keeping with the proposed guideline framework, the basic model designs recognize both fixed and variable costs.

The data source used for the analyses has been the National Center for Education Statistics’ (NCES) Integrated Postsecondary Education Data System (IPEDS). Through IPEDS, NCES surveys approximately 11,000 institutions nationally through a regular data collection cycle on institutional characteristics, student enrollment, staffing, finances, and degrees granted, among others.

The data analysis has focused on two populations of institutions. The first population includes all public colleges and universities in the IPEDS universe, excluding specialized institutions such as stand-alone medical schools (this group totals 1,306). The second population includes all public colleges and universities that are considered "official peers" of Virginia institutions for salary comparison purposes (this group totals 366). The reason for looking at the public peers separately is to determine if there are materially different results from all public institutions as a whole.

The overall methodology employed to date has involved a three-stage process. The initial stage was to run models with various combinations of "cost driver" variables to determine which cost drivers showed up as having the strongest statistical relationships and predictive value in each of the three program areas using the most recent year’s available IPEDS data (1996-97). The results of these initial analyses were shared and discussed with the technical advisory group at the November meeting.

The second stage was to narrow further analysis to those cost driver variables that appeared the most defensible through the results of the initial statistical analyses, as indicated below:

• Academic Support: Unrestricted Instruction, Research, and Public Service Expenditures.
• Institutional Support: Unrestricted E&G Expenditures (Less Institutional Support).
• Student Services: Total Student Headcount.

Also, these analyses were conducted using the two most recent years worth of data available at the time (1995-96 and 1996-97) in order to see if the initial relationships were consistent across both years.

The third stage was to further disaggregate the analyses by institutional type. That is, separate models have been run for the following broad groups of institutions for each of the three support program areas:

• Research
• Doctoral
• Masters/Comprehensive
• Baccalaureate
• Two-year

Institutions were assigned to each of these groups based on their current Carnegie classification. The purpose of this step is to determine if different funding factors might be justified for the different types of institutions in Virginia. For illustration purposes, Exhibit 9 below categorizes each Virginia institution into one of the five groups.

The initial results of these analyses indicated that there are statistically significant differences among the various institutional types that may need to be recognized. Two-year institutions were further disaggregated by FTE enrollment level. The reason for this breakout was to recognize the wide variation in enrollment among two-year institutions, and therefore the possible impact of economies of scale at these institutions. The enrollment ranges were determined based on a graphical analysis of two-year institution data which indicated distinct FTE enrollment level "breakpoints" where economies of scale for the three support program areas appear to haven been realized (generally by the time enrollment levels reach 1,500 FTE students).

The second and third stage results were discussed with the technical advisory group at the January meeting. There was general consensus within the group, based on the regression results, that:

• the appropriate academic support guideline would be academic support as a percent of instruction, research, and public service expenditures (as generated by the base funding guidelines), differentiated by institutional type;
• the appropriate institutional support guideline would be institutional support as a percent of total E&G expenditures less institutional support (as generated by the base funding guidelines), differentiated by institutional type; and
• the appropriate student services guideline would be a fixed plus variable cost model where each institution received a base amount regardless of size, plus a per capita amount to be applied to the institution’s total headcount enrollment, differentiated by institutional type.

Preliminary Recommendation. Based on the analyses conducted to date and the input of the technical advisory committee, MGT’s preliminary base funding guideline methodology recommendations for the three support program areas are outlined in Exhibit 10 below.

Initial regression model results reflecting these preliminary recommendations are included in Appendix C at the end of this report. Further analyses are being conducted in order to refine the guideline rates for each of the three program areas, including the replication of the three relevant regression models using the recently released 1997-98 IPEDS data.

4.3 Operation and Maintenance of Plant

Compared to the support program areas described in the previous section, national data on specific costs related to the Operation and Maintenance of Plant (Plant O&M) is relatively sparse. Thus, our work to date in the area of Operation and Maintenance of Plant has been to search for benchmark data that could be used to develop related funding guidelines for Virginia’s public institutions of higher education. Our initial research found data from a national survey of colleges and universities in 1998-99 that could serve as a basis, which were discussed with the technical advisory group. It was the general consensus of the group that these data would be inappropriate for use in the development of plant O&M guidelines for Virginia institutions, primarily due to the fact that the national data did not reflect the experiences of all types of institutions.

Preliminary Recommendation. There is no recommended approach for plant O&M funding guidelines at this time. Rather, further research will need to be conducted to determine the availability of additional or alternative data sources to inform the development of base funding guidelines for the operation and maintenance of plant. At a minimum, Virginia institutions will need to collect the following data for analysis: total square footage by age of facility and type of construction; total maintained acreage; and utilities expenses.

Over the next several weeks we will continue to work with legislative staff and institutional representatives to refine and calibrate an initial set of base funding guidelines based on the preliminary recommendations outlined in the previous section and to gather all of the necessary input data for each institution. From that point on, the research will involve an iterative process of guideline simulations and further factor refinements to arrive at a final set of recommended base funding guidelines. We envision that the initial set of base funding guidelines and input data could be in place by mid-spring, with a set of final recommendations by early summer.

APPENDICES

APPENDIX A: STUDENT/FACULTY RATIOS BY ACADEMIC DISCIPLINE AND INSTRUCTIONAL LEVEL – SELECTED STATES AND APPENDIX M

Appendix B
Instructional Staffing Level Standards
From Selected Accrediting Organizations

Key: NLNAC – Nursing; AACSB – Business; ABET – Engineering; ABA – Law; CAAHEP – Health;
NAAB - Architecture; NCATE - Education

• Number and utilization of full- and part-time faculty meet the needs of the nursing unit to fulfill its purposes.

• IN.1: The school should provide and manage resources to meet the instructional responsibilities created by the programs offered.

• FD.4.a: The school should maintain a full-time faculty sufficient to provide stability and ongoing quality improvement for the degree programs offered.

• c.)A program at the basic level must have no fewer than three-full-time faculty members (i.e., the fractions of time devoted to the basic-level program by each faculty member must add to at least three.) This statement shall not be interpreted to preclude the accreditation of programs offered primarily by part-time faculty members. The institution must demonstrate that effective mechanisms are in place to assure adequate levels of student-faculty interaction, student advising, and faculty concern for and control over the curriculum, as would be expected in programs offered primarily by full-time faculty members. If the faculty has additional obligations, such as graduate teaching and/or research, additional faculty members must be present to ensure that at least three full-time-equivalent faculty members are devoted to each basic-level program. Under no circumstances should a program be critically dependent on one individual.

• Interpretation 402-2: