LAO Analysis of the 1995-96 Budget Bill
Higher Education Crosscutting Issues

Rethinking the State's Approach to Deferred Maintenance

The current system of funding ongoing maintenance and deferred maintenance creates counterproductive fiscal incentives that encourage the University of California, the California State University, and the California Community Colleges to defer needed maintenance. We recommend specific steps the Legislature should take to resolve the existing backlogs in deferred maintenance and the underfunding of regular maintenance.


To keep the state's facilities at the University of California (UC), the California State University (CSU), and the California Community Colleges (CCC) functional for public use, the state and the systems fund both ongoing maintenance and special repair programs.

"Maintenance" includes (1) janitorial and groundskeeping activities and (2) programs to maintain the condition of facilities and infrastructure/utility systems. "Special repair" refers to maintenance projects that are required periodically and are above the level of expenditures needed for routine maintenance. Examples of special repairs include replacing roofs, painting exteriors, and replacing mechanical/electrical equipment.

Deferred Maintenance: Save Now, Pay More Later

When ongoing maintenance is not sustained at an appropriate level and special repair projects are not accomplished as needed, the result is a backlog of projects termed "deferred maintenance." If repairs to key building and infrastructure components are constantly deferred, facilities can eventually require more expensive investments, such as emergency repairs (when systems break down), capital improvements (such as major rehabilitation), or replacement. Generally, deferral of maintenance projects reduces the useful life of facilities and thus increases future capital outlay needs.

Reported Deferred Maintenance Backlogs Are Huge

Over the past 10 to 15 years, California's three public higher education systems have been in a state of constant maintenance deferral. As a result, the UC estimates that its deferred maintenance backlog exceeds $480 million, of which about $251 million are priority-one projects. (Priority-one deferred maintenance projects are those requiring immediate action to return a facility to normal operation, stop accelerated deterioration, or correct a cited safety hazard.) The CSU estimates that its deferred maintenance backlog exceeds $325 million, of which about $108 million are priority-one projects. The CCC Chancellor's Office estimates that the statewide community college deferred maintenance backlog is about $90 million.

These figures represent each segment's evaluation of "need." Based on our campus visits, we believe that the total deferred maintenance backlog is in the range of several hundred millions of dollars; however, our review indicates that the specific magnitude of the problem is uncertain for three reasons:

Budget History and 1996-97 Proposals

UC and CSU.In 1994-95 and 1995-96, the annual Budget Act and related legislation authorized loan financing for "priority-one" deferred maintenance projects at the UC and CSU that would have an anticipated useful life of at least 15 years. The state provided loans of $17 million to $25 million per segment per year. These loans are being repaid through augmentations to the UC and CSU General Fund budgets.

Although the 1996-97 budget proposes a change in funding source for UC and CSU deferred maintenance--no loans this year--it essentially continues the previous years' approach. That is, it augments the segments' operating budgets by an amount that is small in comparison to the total amount of the deferred maintenance backlog. Specifically, the budget allocates about $10 million each to UC and CSU for high-priority deferred maintenance. The UC amount is from one-time general obligation (GO) debt financing and the CSU amount is from General Fund monies (with the intent that this is a "base" adjustment, which would be available annually).

Community Colleges.For the last several years, the state has provided deferred maintenance funding of $8.7 million annually to the community colleges from Proposition 98 funds. Due to a required dollar-for-dollar local match, this annual appropriation generates about $17 million in deferred maintenance activities.

The budget proposes a total of $26.2 million for community college deferred maintenance in 1996-97. It proposes to spend $17.5 million in one-time 1995-96 Proposition 98 funds for CCC deferred maintenance, and waive the local match requirement for these funds. The budget also proposes the ongoing amount of $8.7 million from 1996-97 funds, but maintains the local match requirement for these funds.

Problems With the State's Current Approach

Figure 6 (see next page) summarizes the major shortcomings of the state's current approach to deferred maintenance. Most importantly, the state's current approach treats deferred maintenance as an ongoing "program." The existence of deferred maintenance, however, really represents a maintenance program failure. A deferred maintenance project is one that should have been addressed in a prior year under a properly functioning regular maintenance program.

One reason for the failure of the segments' regular maintenance programs is simple--regular, ongoing maintenance has been underfunded. Both the state and the segments have contributed to this underfunding. Moreover, separate funding for deferred maintenance may actually create a fiscal incentive to defer projects rather than deal with them in a more timely manner. Below, we discuss these problems in detail.

Figure 6
Problems in the State's Current Approach to Maintenance
The Current Approach:
Deferred maintenance is a state-funded program. State provides relatively small annual augmentations for this purpose to the systems' base operating budgets.
  • Does not address underlying causes of deferral:
  • - State underfunds regular maintenance programs.
    - Segments not held accountable for spending state maintenance funds for that purpose.
  • Counterproductive fiscal incentive--makes projects less expensive for systems to address as deferred maintenance than under a regular maintenance program.
  • State Funding: Maintenance Has Been a Low Priority.Underfunding of maintenance has occurred in part because the state did not budget sufficient funds to maintain both student instruction and maintenance and special repairs. Maintenance has been viewed as a lower priority than the need to maintain the quantity and quality of direct student instruction. It has been seen as more discretionary and, therefore, deferrable. As a result, spending on maintenance has lagged, and facilities have prematurely deteriorated. In the short-run, this policy has mitigated the effect of the recession's low-revenue years on higher education enrollments and the quality of instruction. If pursued in the long-run, however, it would constrain futureenrollment and quality levels as the state would eventually have to redirect funds otherwise available for these purposes to repair and replace prematurely worn out buildings and infrastructure.

    Segments Not Held Accountable.Underfunding of regular maintenance has also occurred in part because the segments redirected funds budgeted by the state for routine maintenance to other activities. This is because there is no framework under which the state holds the systems accountable for the outcomes of their maintenance programs.

    The state has the primary responsibility for funding maintenance at the systems. It has little control, however, for determining ongoing maintenance spending at each campus. Although the state has periodically reviewed some specific maintenance and repair-related issues (such as whether the UC and CSU maintenance cost standards are consistent), there are serious gaps in oversight. For example, the UC and CSU maintenance standards have not been reviewed since the mid-1980s, and there is no systematic process for reviewing where actual spending levels stand comparison to the standards. As a result, the UC and CSU have significant flexibility in determining the level of maintenance that actually occurs at each campus. Similarly, state funds are allocated to the CCCs on the basis of maintenance and operation workload, but the colleges have virtually unlimited discretion in determining what kinds of maintenance these funds support--or whether they are used to support maintenance at all.

    Fiscal Incentive to Defer Makes a Bad Situation Worse. The state's current method of funding deferred maintenance actually provides an incentive for the systems to defer projects. This is because the state has addressed the maintenance problem primarily by adding state monies for deferred maintenance over and above the regular operating budget of the systems. As a result, the current funding arrangement rewards the systems for maintenance deferrals by providing a higher level of funding for deferred maintenance.

    For the CCCs, the state provides matching funds under the deferred maintenance program. From the colleges' perspective, therefore, it costs less--in terms of system discretionary funds that must be used--to address a repair under the state-funded deferred maintenance program than it does to address it under a regular maintenance program. Thus, the fiscal incentive offered by the state's approach points in the wrong direction.

    New Approach Needed in 1996-97

    We recommend that the Legislature: (1) increase funding for ongoing maintenance and hold the systems accountable for better results, (2) prohibit the addition of any new projects to existing deferred maintenance backlogs, and (3) start a process to eliminate the existing backlogs.

    We believe that the improvement in the state's economic and budgetary situation makes this a good time for the state to begin resolving the maintenance problems at the UC, CSU, and CCC. We recommend the Legislature follow the principles outlined in Figure 7 (see next page) as it considers this issue. Figure 7 also summarizes our recommendations to the Legislature for putting these principles into action. Below, we discuss these recommendations primarily as they apply to the UC and CSU. We present a detailed proposal for the community colleges later in our analysis of the CCC budget.

    Figure 7
    Principles and Recommendations for Reform of
    Higher Education Maintenance Funding
  • Adequately fund regular, ongoing maintenance.
  • Hold the systems accountable for improving regular maintenance efforts.
  • Shift fiscal incentives to discourage deferral of projects.
  • Augment the segments' maintenance budgets in 1996-97.
  • Require a segmental funding match and increased maintenance efforts.
  • Require all funds budgeted for maintenance to be spent for that purpose.
  • Prohibit adding new projects to existing deferred maintenance backlogs after January 1, 1996.
  • Reject debt financing of UC deferred maintenance in 1996-97.
  • Develop a plan to eliminate existing deferred maintenance backlogs over time.
  • Provide Adequate Maintenance Funding

    The first step in correcting the deferred maintenance problem is ensuring the segments adequately fund ongoing maintenance. The segments report that their maintenance budgets are currently underfunded relative to state standards. Specifically, the UC and CSU advise that their building and infrastructure maintenance budgets (which exclude custodial and grounds maintenance) are at least $33 million and $22 million below the standard, respectively.

    Given that this shortfall is the responsibility of both the state and the segments, we recommend that the state and the segments share the burden of restoring maintenance funding to adequate levels. Specifically, we recommend:

    State Augmentation.We recommend the Legislature provide a General Fund augmentation of $10 million each to the UC and CSU and $25 million for the CCC. We believe the increase in state funding would be a step towards ensuring that the segments provide an adequate level of maintenance, thereby avoiding higher deferred maintenance costs later.

    Segment Match.We recommend the Legislature require the segments to match from internal sources the augmentation provided by the state. Thus, the total increase in maintenance funding for each segment would be $20 million for UC and CSU, and $50 million for the CCC, an amount that would move the segments toward sufficient maintenance funding. In order to match state maintenance funds, we recommend the Legislature redirect funding from other priorities, as shown below:

    Accountability. The Legislature should require the segments to use all state funding budgeted for maintenance--plus the proposed segment match--solely for the purpose of maintenance. The Legislature should define "maintenance" in this regard as efforts to maintain facilities and infrastructure, as opposed to janitorial services and groundskeeping. While the latter are important, they have no major effect on the length of facilities' useful life. During budget hearings, we will recommend Budget Bill language necessary to accomplish these recommendations.

    Cap the Backlog at Its Current Level

    The second step in correcting the deferred maintenance problem is to hold the segments responsible for any new deferred maintenance costs in the future. Given increased funding for regular maintenance, the systems should commit to the proper maintenance of all existing facilities. The Legislature should make it clear that it will not fund projects that are deferred in the future.

    This means the state needs to refine the segments' lists of existing deferred maintenance projects to which no new projects could be added. After the Legislature closes the segments' existing lists, it should ensure that state control agencies have an opportunity to review the projects on the list and determine whether they are appropriately classified as deferred maintenance (as opposed to capital renovation, for example).

    We therefore recommend the following supplemental report language, which would require the segments to submit their lists of deferred maintenance projects as of January 1, 1996, to the Legislative Analyst's Office (LAO) and the Department of Finance (DOF). The lists would be reviewed and evaluated by the LAO, the DOF, and representatives of the segments:

    The University of California Office of the President, the Chancellor of the California State University System, and the California Community Colleges Chancellor's Office shall submit to the Legislative Analyst's Office (LAO) and the Department of Finance (DOF) upon enactment of the 1996 Budget Act, the list of segment-wide deferred maintenance projects identified as of January 1, 1996. The LAO, DOF, and representatives of the UC, CSU, and CCCs shall review the projects for merit--based on criteria agreed to by the parties. Based on this process, the LAO and the DOF shall jointly present--no later than November 1, 1996--a final list of all existing deferred maintenance projects at California public postsecondary education institutions.

    It is the intent of the Legislature to provide deferred maintenance funding in the future only for those projects included on the list presented on November 1, 1996. It is further the intent of the Legislature that the segments shall not defer maintenance in the future. Given the increased level of funding for regular maintenance provided in the Budget Act, and the Legislature's intent to fully fund regular maintenance in future years, the Legislature regards any deferred maintenance project that is not included on the November 1, 1996 list as the fiscal responsibility of the segment, not of the state.

    Reject 1996-97 Debt Financing of UC Deferred Maintenance Projects

    We recommend the Legislature reject the proposal to provide $10 million from proposed general obligation bond funds for UC deferred maintenance in 1996-97 (delete Item 6440-001-0658). We believe that the use of debt financing for deferred maintenance projects prior to the thorough review of UC's existing deferred maintenance list (as described above) would be premature. Moreover, this one-time, relatively small, amount of deferred maintenance funding in the budget year leaves future funding for UC's deferred maintenance backlog an open question. Below we discuss some funding sources--including debt financing--that could be used as part of a long-range plan to reduce the backlog of deferred projects identified in the proposed November 1, 1996 report to the Legislature. In 1996-97, the UC should use funds from its regular operating budget to address any urgently needed deferred maintenance.

    Develop a Plan to Eliminate theExisting Deferred Maintenance Backlog

    The third step in resolving the deferred maintenance problem is to develop a way to fund the existing deferred maintenance backlog. The Legislature, working with the administration and the systems, should specify a time period--of probably five to ten years--to eliminate the current backlog of projects. The amounts addressed each year should be included in the annual Budget Act--beginning with 1997-98--under a separate deferred maintenance item for each segment.

    Potential Funding Sources.To ultimately eliminate the current deferred maintenance backlog, the following sources should be considered:

    We think it is appropriate to select a combination of the above (all of which would be temporary in nature) in order to provide as much funding as possible to quickly eliminate the deferred maintenance backlog. Given that the state and the segments each bear some responsibility for the deferral of these projects, we recommend that the Legislature require the segments to match any state funds allocated to reduce the backlog.


    A long-run strategy to address maintenance failures at the state's higher education segments is essential to protect the state's investment in higher education buildings and infrastructure. Unless the state acts now to (1) bring the systems' maintenance spending to adequate levels and (2) hold the systems' accountable for addressing their ongoing regular maintenance needs, maintenance will continue to be deferred. As a result, the state will face higher future costs of renovating and replacing prematurely worn out facilities.

    We recognize that other state agencies and the K-12 schools also have significant deferred maintenance backlogs. We believe, however, that the state should start by addressing the higher education problem because the size of the deferred maintenance backlog in higher education significantly exceeds the combined total of deferred maintenance needs in other state agencies.

    UC and CSU Faculty Salaries: Update

    Consistent with legislative intent, the California Postsecondary Education Commission staff have consulted with a technical advisory committee regarding the faculty salary methodology and plan to recommend to the commission that the committee's proposed compromise methodology be adopted.

    Every year, pursuant to Senate Concurrent Resolution 51 of 1965, the California Postsecondary Education Commission (CPEC) submits to the Governor and the Legislature an analysis of faculty salaries at the University of California (UC) and the California State University (CSU) compared to the average salaries of specified groups of higher education institutions. The UC and CSU use the average salaries of their respective comparison schools to determine the reasonableness of the faculty salaries they provide. The CPEC's analysis is based on a set of procedures and calculations that are collectively referred to as the "faculty salary methodology."

    Figures 8 and 9 (see next page) display the faculty salary comparison institutions for UC and CSU, respectively.

    In our Analysis of the 1995-96 Budget Bill, we noted that the last comprehensive review of the faculty salary methodology occurred roughly a decade ago. We also found that the existing methodology resulted in the calculation of faculty salary gaps--the amounts that average faculty salaries at UC and CSU fall below the salaries paid at their comparison institutions--that are too large, particularly at the CSU. This was primarily because the methodology did not adequately account for the significantly higher proportion of full professors at the CSU compared to its 20 comparison institutions. The UC's faculty staffing patterns are more in line with its eight comparison institutions.

    To address this issue, we recommended the Legislature use a faculty salary methodology based on a comparison of the simple average CSU and UC faculty salaries with the simple average faculty salaries of their respective comparison institutions. We also suggested reviewing CSU's comparison institutions list because it contains at least one major doctoral-granting institution (the University of Southern California). Universities that grant significant numbers of doctoral degrees generally pay higher faculty salaries than institutions with teaching missions that are more similar to the CSU. (Under the state's Master Plan for Higher Education, the CSU is authorized to offer doctoral degrees only through joint arrangements with the UC or private universities.) For this reason, we questioned whether such schools should be included in CSU's comparison group.

    Figure 8
    The University of California
    Comparison Institutions for Faculty Salaries
    Public Institutions:Private Institutions:
    University of Illinois (Champaign-Urbana)Harvard University
    University of Michigan (Ann Arbor)Massachusetts Institute of Technology
    University of Virginia (Charlottesville)Stanford University
    State University of New York (Buffalo)Yale University

    Figure 9
    The California State University
    Comparison Institutions for Faculty Salaries
    Public Institutions:University of Maryland (Baltimore County)
    Arizona State UniversityUniversity of Nevada (Reno)
    Cleveland State UniversityUniversity of Texas (Arlington)
    George Mason UniversityUniversity of Wisconsin (Milwaukee)
    Georgia State UniversityWayne State University
    Illinois State UniversityPrivate Institutions:
    North Carolina State UniversityBucknell University
    Rutgers University (Newark)Loyola University of Chicago
    State University of New York (Albany)Reed College
    University of Colorado (Denver)Tufts University
    University of ConnecticutUniversity of Southern California

    The Legislature, per our recommendation, specified its intent in the Supplemental Report of the 1995 Budget Act that the CPEC, in consultation with a technical advisory committee, make recommendations on the appropriateness of addressing the above issues to the Legislature, the Department of Finance (DOF), and the Legislative Analyst's Office (LAO) by December 1, 1995.

    Group Recommends New Methodology.During fall 1995, the CPEC convened a technical advisory group to review the faculty salary methodology composed of representatives from the UC and CSU administration and faculty, the DOF, the LAO, and other interested parties. In addition to the issues noted above, the systems raised two additional concerns:

    Ultimately, members of the technical advisory committee developed a compromise faculty salary methodology. Under the compromise, the methodology would be changed to: The proposed compromise would maintain the existing comparison institutions for the UC and CSU, reflecting a prevailing view that the comparison institution lists appear reasonable when viewed as a whole. The representatives of the technical advisory group also suggested to the CPEC staff that the proposed methodology take effect in 1996-97. The group suggested another review of the methodology in time for consideration of the 1999-2000 budget, which would review the appropriateness of using the simple average faculty salaries to measure the salary gaps at UC and CSU and the appropriateness of the institutions on the CSU's comparison list.

    We believe the proposed compromise faculty salary methodology is a reasonable one to use through 1998-99. It is also important, however, to review the methodology periodically to ensure that issues of interest to the Legislature are thoroughly considered. For this reason, we will continue to review the issue of comparability and the use of comparison institutions to set faculty salaries.

    Under the existing methodology, the CPEC identified faculty salary gaps of 10.4 percent for the UC and 12.7 percent for the CSU in 1995-96 (before any faculty salary increases). If the proposed compromise methodology is adopted for 1996-97, there would continue to be a 10.4 percent gap for the UC, but the CSU gap would decline to 9.5 percent. At the time of this analysis, the CPEC staff were anticipating that they would recommend to the commission that the suggested compromise be adopted. The commission will act upon its staff recommendations on the faculty salary methodology later in spring 1996.

    No Budget-Year Implications. As a practical matter, the issues we discuss above are not likely to have implications for the 1996-97 budget. Specifically, the UC budget proposes a 5 percent general faculty salary increase, plus merit salary adjustments for eligible faculty. The CSU budget proposes an overall 4 percent compensation increase, with specific allocations subject to collective bargaining. Even with these proposed increases, there would still be a faculty salary gap of about 5.4 percent at UC and CSU. (The actual CSU gap would depend on the results of collective bargaining.)

    Marginal Cost at UC and CSU: Update

    Proposed marginal cost calculations for the UC and CSU have recently been developed in line with legislative intent.

    Through 1991-92, the relationship between the budget and enrollment for the University of California (UC) and the California State University (CSU) was clearly defined. As enrollments changed, dollars were added or subtracted from the UC and CSU budgets based on agreed-upon cost calculations (called "marginal cost," as defined below) for serving new students in each of these systems. Beginning in 1992-93, this relationship was suspended. Faced with significant General Fund reductions and student fee increases, the Governor and the Legislature deleted Budget Bill language in the 1992 Budget Act that provided funding adjustments in the event of certain fluctuations (up or down) in enrollment.

    In the Supplemental Report of the 1994 Budget Act, the Legislature expressed its intent that, beginning with the 1996-97 budget, the state return to the use of marginal cost as the basis for funding enrollment changes and that representatives from the UC, CSU, the Department of Finance (DOF), and the Legislative Analyst's Office (LAO) review the components of the marginal cost calculation and propose any modifications in the development of the 1996-97 budget. The Legislature also re-adopted the pre-1992 Budget Act language in the 1994 Budget Act to provide funding adjustments if actual enrollments are more than 2 percent above or below the enrollment targets. However, the Governor vetoed the language, stating that adjustments to the UC and CSU funding levels "should be addressed through additional legislative action when compared to other essential financial needs." (As a practical matter, actual UC and CSU enrollments since that time have not varied by more than 2 percent in any one year compared to the Legislature's enrollment targets for these systems, and thus the veto of this language has had no specific budget implications.)

    Below, we define "marginal cost" and review recent actions that respond to the Legislature's intent that the calculation be reviewed in time for consideration of the 1996-97 budget.

    What Is "Marginal Cost"?"Marginal cost" of a university education is generally used to describe the various per-student costs of serving relatively small blocks of new students (such as 1,000). The marginal cost is often less than the average cost because it reflects what are called "economies of scale"--that is, certain fixed costs (such as for central administration) may change very little as new students are added to an existing campus. The major components of a marginal cost calculation generally include faculty costs, library services, student services (such as counseling and financial aid), and certain other administration costs. The marginal costs of a UC and CSU education are funded by the state General Fund and student fee revenues.

    Issues Related to Prior Marginal Cost Calculation.In response to the Legislature's directive, representatives from the UC, CSU, DOF, and LAO reviewed the marginal cost calculations that were in effect in 1991-92 and potential changes to them. The major issues raised about the 1991-92 marginal cost calculations generally fell into the following two categories:

    The representatives, however, also noted that many parts of the marginal cost calculation continue to be reasonable. These include (1) determining the marginal cost for the budget year based on current-year costs and (2) setting the additional cost of hiring faculty to serve additional students at entry-level, rather than average, salaries.

    Overview of Proposed New Marginal Cost Calculations.The four agencies ultimately agreed to specific marginal cost calculations for the UC and CSU. The proposed calculations for each of the systems:

    As was the practice through 1991-92, the marginal cost for each budget year would be updated to reflect estimated current-year expenditures. Finally, the General Fund and student fee revenues would support the total marginal cost in relation to the past year's "split" between the two funding sources (for example, the approximate 1995-96 "split" would be used as the basis for the 1996-97 calculation). The split is needed to calculate the General Fund amount that must be added for the support of new students.

    Under the proposed calculations, the marginal cost identified for UC in 1996-97 would total $8,730 ($6,809 from the General Fund and $1,921 from student fee revenues). The marginal cost identified for the CSU for 1996-97 would total $5,917 ($4,734 from the General Fund and $1,183 from student fee revenues).

    Given changes in budgeting practices that occurred over time in both segments, it is difficult to compare the proposed marginal costs to those that existed in 1991-92. For the UC, our review of the General Fund portion of the marginal cost indicates that the proposed 1996-97 amount ($6,809) is $809, or about 14 percent above the 1991-92 level of $6,000. This increase is in line with the cumulative impact of inflation over the time period. For the CSU, the total marginal cost in 1992-93 (the most recent year in which a detailed calculation is available) was about $4,500, compared to a total proposed 1996-97 amount of $5,917. However, it is not meaningful to compare the two figures because the 1992-93 figure excluded significant costs, such as costs for some student services.

    Conclusion. There is no particular "correct" way to calculate marginal cost. We believe, however, that the marginal cost agreement described above is a reasonable one and it also treats the UC and CSU in a consistent manner. The CSU's proposed 1996-97 budget plan would fund anticipated enrollment increases based on the proposed marginal cost. The UC indicates that it intends to implement the proposed marginal cost as soon as possible, but not later than 1997-98. The systems' plans appear reasonable in light of the Legislature's intent that the state return to the use of marginal cost as the basis for funding enrollment changes.

    Return to LAO Budget Analysis Table of Contents

    Return to LAO Home Page