Deferred Maintenance Backlog Statistics: Federal and Higher-Ed Data

Deferred maintenance is the practice of postponing repairs and upkeep to a later date, usually to meet a budget constraint. For the definition, the calculation of the Facility Condition Index, and the business case for funding a backlog, see our companion article, How Deferred Maintenance Becomes a Costly and Risky Burden. This page does something different: it collects the widely cited backlog figures, names where each one actually comes from, and flags where the numbers are less solid than they look.

The pattern here is the same one that runs through our other reference pages. A figure gets repeated across vendor blogs and trade press until it reads like an established fact, and the useful work is tracing it back to the document that produced it and stating plainly how it was measured.

Federal deferred maintenance backlog

The federal government owns roughly 277,000 buildings, and its annual operating and maintenance costs for them exceeded $10.3 billion in FY2023, according to the Government Accountability Office. Managing federal real property has been on GAO’s High-Risk List for 22 years. In its 2025 High-Risk update, GAO added building condition as a specific high-risk topic, citing large increases in the cost of addressing deferred maintenance.

The headline federal figure is GAO’s roll-up across the Department of Defense and federal civilian agencies. GAO reported that these building deferred maintenance and repair backlogs more than doubled, from $171 billion to $370 billion, between FY2017 and FY2024. GAO’s stated concern is that unless the trend reverses, assets will deteriorate to the point of needing premature replacement, which is generally more expensive than the delayed repairs would have been.

Individual agency figures fill in the picture:

Table 1: Reported federal deferred maintenance backlogs. Figures are drawn from Government Accountability Office reports, Congressional Research Service products, National Park Service data, and the NASA Office of Inspector General. They are absolute dollar backlogs and are not normalized to portfolio size, so they should not be compared directly against one another as measures of relative condition. The DoD figure is included in the government-wide total, so the two should not be added together.

National Park Service

The National Park Service backlog is a high-profile federal deferred maintenance figure, in part because it has drawn sustained congressional attention. The Park Service’s own Federal Real Property Profile data put the estimate at $23.258 billion for FY2023 and $24.237 billion at the end of FY2025. The intervening year does not sit on a straight line: the FY2024 estimate was $22.988 billion, a dip before the FY2025 rise. The Congressional Research Service, using the figure reported at the time ($23.263 billion for FY2023), calculated that the estimate grew over the decade from FY2014 to FY2023 by about 102 percent in nominal dollars and about 38 percent in inflation-adjusted dollars.

Two things about that growth are worth stating clearly. First, the Great American Outdoors Act established the National Parks and Public Land Legacy Restoration Fund, which provided up to $1.33 billion annually to NPS for each of FY2021 through FY2025, with FY2025 the final authorized deposit year. So the backlog grew even while a dedicated multi-year fund was being spent against it. Second, part of the movement in the estimate came from changes in how NPS calculates the backlog rather than from physical deterioration alone.

That estimation point has a specific and cautionary example behind it. In a September 2023 report (Report No. 2020-CR-066), the Department of the Interior Office of Inspector General found that the NPS deferred maintenance estimate rose from $11.3 billion in FY2016 to about $20 billion in FY2021 while the number of identified assets stayed relatively constant. The OIG found that NPS had applied a blanket 35 percent markup to its FY2021 estimate for non-transportation assets (transportation assets such as paved and unpaved roads and bridges already carried a 35 percent project-execution markup). That markup added roughly $3.7 billion and brought the reported FY2021 total to about $23.7 billion. The OIG found insufficient documentation to show the markup amount was reasonable, and it noted that the Interior policy NPS cited as justification contained no reference to a standard 35 percent figure. NPS disagreed with the OIG’s characterization. The takeaway for anyone citing the NPS number is that it is a real and heavily scrutinized figure, and that its year-over-year movement reflects methodology as well as physical condition.

Department of Defense

DoD manages one of the largest real property portfolios in the federal government: more than 700,000 facilities with a combined replacement value of about $2.2 trillion as of FY2023. GAO reported that DoD’s deferred maintenance backlog stood at $181.1 billion for FY2023, drawn from DoD’s own FY2023 Agency Financial Report and excluding military family housing. That figure is up from the $137 billion DoD reported for FY2020, consistent with the broader upward trend across federal real property. Over the five years preceding the FY2023 report, DoD reported spending an average of about $15.3 billion a year to maintain and repair existing facilities and about $14.6 billion a year to build new ones. GAO separately reported that DoD funded roughly 80 percent of its facility sustainment requirement, below its 90 percent goal.

NASA

NASA manages about $40 billion in facility assets across more than 5,000 buildings and structures, and the NASA Office of Inspector General reported that more than 75 percent of that constructed infrastructure is beyond its design life. The OIG put NASA’s deferred maintenance backlog at $2.66 billion as of 2020, against a Construction of Facilities budget of roughly $357 million per year. The OIG also stated that deferring maintenance had pushed work into unscheduled repairs that can cost up to three times more than scheduled maintenance would have. That multiplier is the OIG’s own characterization for NASA’s facilities; readers looking at the general reactive-versus-planned cost question can see our separate treatment of that claim in Reactive Maintenance Costs 3 to 5x More: Tracing the DOE Claim.

Higher education deferred maintenance

A widely cited source for campus deferred maintenance is Gordian’s annual State of Facilities in Higher Education report. Gordian is a commercial provider of facilities and construction cost data, and the report is drawn from the company’s own database, which it describes as covering 43,000 campus buildings and 1.1 billion gross square feet across North America. That vendor origin does not make the figures wrong, and it is worth noting because the report is a large and widely used benchmarking dataset in the sector. We label it as vendor-conducted for the same reason we label every proprietary dataset on this site.

Gordian’s 2026 report (its 13th edition) put the deferred capital renewal burden at $156 per gross square foot, an 8 percent increase over the prior year and roughly double where it stood around 2008. The report also found that institutions were investing only 73.5 percent of the funding needed to keep the backlog from growing, and that operating budgets ran 18.5 percent below target. One precision point matters here: the $156 figure is deferred capital renewal per gross square foot, not a total dollar backlog and not a Facility Condition Index. It describes accumulated renewal need for major building systems, normalized by floor area.

For a national dollar figure, the widely cited number traces to earlier Gordian and APPA work. An April 2021 Gordian and APPA analysis estimated higher education faced roughly $112 billion in urgent deferred renewal nationally, against annual spending of about $36.8 billion on operations, maintenance, and utilities and about $27 billion on capital investment in new and renovated space. That estimate was carried into Gordian’s 9th-edition report, released in March 2022. The $112 billion figure is now several years old and predates the more recent per-square-foot increases, so it is best cited with its date attached rather than presented as a current total.

Facility Condition Index: what the benchmark actually measures

The Facility Condition Index (FCI) is the metric that lets deferred maintenance be compared across buildings of different sizes. It is calculated as the cost of a building’s identified deficiencies divided by the cost to replace that building, expressed as a percentage, so a lower FCI indicates better condition. Our companion article walks through the calculation and the replacement-value denominator in detail.

The FCI was developed by the firm Applied Management Engineering and published in 1991 in Managing the Facilities Portfolio, a NACUBO volume by Applied Management Engineering and Sean C. Rush. That book is generally credited as the first place the index appeared in print, and FCI has since become a widely used facilities-condition metric.

The complication is on the interpretation side. FCI is routinely presented with “good, fair, poor” threshold bands described as an industry standard, but those bands are framework-specific rather than universally standardized, and different sources publish different cutoffs. As one example of how source-specific the bands are, a 2009 GAO study of school facilities adopted a scale developed for Montana’s statewide school assessment (an FCI below 10 percent as good, 10 to 19 percent as fair, and 20 percent and above as poor) rather than any single national standard. GAO cautioned in that 2009 study that FCI values are useful for comparing facilities only when they are calculated using a consistent methodology. Any specific set of bands should be confirmed against your own governing framework, not treated as fixed.

A note on replacement-value terminology

The denominator underneath FCI and underneath most maintenance-budgeting ratios goes by several names, and the names are used inconsistently across the maintenance and facilities worlds. This can complicate cross-industry comparisons, so it is worth stating plainly.

Facilities and asset-management frameworks use several related replacement-value terms, including facility replacement value (FRV), replacement asset value (RAV), current replacement value (CRV), and plant replacement value (PRV). They describe broadly similar replacement-cost concepts, the estimated cost to replace the asset base with equivalent new assets at today’s prices, but they should not be assumed to be interchangeable, because asset scope, inclusions, valuation methods, and update practices can differ between frameworks. FRV and CRV are the terms more common in facilities and higher-education work. RAV is the term more common in plant maintenance and reliability, where it appears as the denominator in metrics such as maintenance cost as a percentage of RAV (see our maintenance and reliability glossary). PRV appears frequently in higher-education facilities reporting. When a figure will be benchmarked externally or written into a budget justification, use the definition from the governing framework rather than a generic one.

Why these figures vary and how to use them

Three structural things make deferred maintenance figures hard to compare, and being explicit about them is more useful than picking a single number to headline.

The first is normalization. Federal agency figures are absolute dollar backlogs. Gordian’s headline higher-education figure is dollars per gross square foot. FCI is a ratio against replacement value. All three are legitimate, and none converts cleanly into the others without information the published summaries usually omit.

The second is methodology change over time. The National Park Service example is a clear documented case, where a large share of a reported increase came from an estimation markup rather than new physical deterioration. Backlog figures that move sharply year over year deserve a look at whether the estimation method changed before the movement is read as real-world decline.

The third is data provenance. Federal figures come from GAO audits and evaluations, inspector-general reporting, Congressional Research Service analysis, and agency financial data. The widely cited higher-education figure comes from a single vendor’s proprietary database. Both are useful, and they carry different kinds of authority, which is worth signaling whenever the numbers are cited side by side.

The defensible practice, the same one we recommend for every benchmark on this site, is to cite each figure with its source and its measurement basis attached, to note when a number reflects a methodology change, and to use your own facility condition assessment as the baseline for any decision rather than a national headline figure.

Frequently Asked Questions

What is a deferred maintenance backlog?

A deferred maintenance backlog is the accumulated cost of repair and upkeep work that has been identified as needed but postponed to a later date, usually because of budget constraints. In facilities and real-property contexts it is typically expressed as a total dollar figure, as a cost per square foot, or as a Facility Condition Index ratio against replacement value. For a fuller definition and the business case for addressing a backlog, see our companion article on deferred maintenance.

How big is the National Park Service maintenance backlog?

The National Park Service reported a deferred maintenance and repair backlog of $23.258 billion for FY2023 and $24.237 billion at the end of FY2025, based on Federal Real Property Profile data. The estimate grew over the prior decade both from cost inflation and new needs and from changes in how NPS calculates the backlog, including a 35 percent markup applied to non-transportation assets in FY2021 that the Department of the Interior Office of Inspector General found insufficiently documented.

What is the total federal deferred maintenance backlog?

The Government Accountability Office reported that federal building deferred maintenance and repair backlogs across the Department of Defense and federal civilian agencies more than doubled from $171 billion to $370 billion between FY2017 and FY2024. GAO added federal building condition to its High-Risk List in 2025 in part because of that increase.

What is the deferred maintenance backlog in higher education?

Gordian’s 2026 State of Facilities in Higher Education report put deferred capital renewal at $156 per gross square foot, an 8 percent year-over-year increase and roughly double its level around 2008. This figure is drawn from Gordian’s proprietary database of campus facilities and is a per-square-foot renewal figure rather than a total dollar backlog. An earlier Gordian and APPA analysis put national urgent deferred renewal at roughly $112 billion in 2021.

What is a good facility condition index (FCI) score?

FCI is the cost of a building’s deficiencies divided by its replacement value, expressed as a percentage, so lower is better. Published “good, fair, poor” thresholds are framework-specific rather than universally standardized, and different sources use different cutoffs. GAO cautioned in a 2009 study that FCI values are useful for comparison only when they are calculated using a consistent methodology, and that is the safer principle to follow, so any specific cutoff should be confirmed against your own governing framework rather than treated as a fixed standard.

Why do deferred maintenance figures vary so much between sources?

Figures vary for three main reasons. They use different normalizations, with some reported as absolute dollars, some as dollars per square foot, and some as a ratio against replacement value. They change when estimation methods change, as the National Park Service example shows. And they come from different kinds of sources, with federal figures produced by government audits and financial reporting and the widely cited higher-education figure produced by a single vendor’s proprietary database. Citing each figure with its source and measurement basis attached is the only reliable way to compare them.

Sources

• Government Accountability Office, Federal Real Property: Disposing of Unneeded Facilities Could Help Reduce Maintenance Backlog (GAO-25-108400), April 2025.
• Government Accountability Office, DOD Real Property: Actions Needed to Improve Oversight of Underutilized and Excess Facilities (GAO-25-106132), March 2025.
• Government Accountability Office, Defense Infrastructure: DOD Should Better Manage Risks Posed by Deferred Facility Maintenance (GAO-22-104481), January 2022.
• Government Accountability Office, School Facilities: Physical Conditions in School Districts Receiving Impact Aid for Students Residing on Indian Lands (GAO-10-32), October 2009.
• Congressional Research Service, National Park Service (NPS) Appropriations: Ten-Year Trends (R42757).
• Congressional Research Service, National Park Service Deferred Maintenance: Overview and Issues (R48136), 2024.
• National Park Service, Infrastructure “By the Numbers,” FY2025 Federal Real Property Profile data (nps.gov).
• Department of the Interior Office of Inspector General, The National Park Service Faces Challenges in Managing Its Deferred Maintenance (Report No. 2020-CR-066), September 2023.
• NASA Office of Inspector General, NASA’s Construction of Facilities (IG-21-027), September 2021.
• Gordian, State of Facilities in Higher Education, 13th edition, 2026.
• Gordian and APPA, State of Facilities in Higher Education, 9th edition, March 2022; national estimate from the April 2021 Gordian/APPA analysis.
• APPA, Body of Knowledge, Deferred Capital Renewal and Deferred Maintenance (citing Applied Management Engineering and Sean C. Rush, Managing the Facilities Portfolio, NACUBO, 1991).