Capital Planning & Reserve Studies for Buildings

Every building is spending capital right now. The only question is whether it is being spent on a schedule the owner chose or on a schedule the roof chose. A chiller that fails in July, a parking deck that fails its inspection, an elevator that finally stops passing — none of those are surprises to the building. They are only surprises to the budget.

Capital planning is the discipline of moving that spending from the second category into the first. It produces a document — call it a capital plan, a capital needs assessment, or a reserve study depending on who is asking — that says what major expenditures a property faces over the next ten to thirty years, roughly when, roughly how much, and where the money comes from. This guide covers how one is actually built, how the funding math works, and what makes the difference between a plan that gets used and a binder that gets filed. It is the work we do as capital planning for owners and boards across Northern Virginia, DC, and Maryland.

What is a capital plan, and how is a reserve study different?

They overlap enough to be confused and differ enough to matter.

A capital plan is an owner’s forecast of major expenditures — replacements, modernizations, and large repairs — laid out year by year with estimated costs. It is a management and budgeting tool. Its audience is whoever approves the budget, and its output is a decision about what gets funded next year and what waits.

A reserve study is a more formalized instrument used primarily by community associations — condominiums, HOAs, co-ops — and it has a standard structure. In the Community Associations Institute’s Reserve Study Standards (opens in a new tab), a reserve study has a physical analysis (a component inventory, a condition assessment, and life and valuation estimates) and a financial analysis (the current fund status and a recommended funding plan). Its audience includes the board and the owners, and in some states, such as Maryland, a statute sets what it must contain; its output is a recommended contribution rate.

A third relative shows up in transactions: the property condition assessment, commonly scoped to ASTM E2018 (opens in a new tab) (the current edition is E2018-24, as of October 2026), produced during due diligence. It sorts opinions of cost into immediate costs and short-term costs. A table of long-term costs, over an evaluation period the user sets, is optional under the standard, and lenders can require one. Buyers commission them; owners inherit them; and an inherited PCA is often the best starting point a new commercial owner has for a real capital plan. ASTM itself says those long-term costs are not suitable for setting reserves for condominiums and other common-interest properties.

The distinction that matters operationally is this: a reserve study answers how much should we be setting aside, and a capital plan answers what are we doing and when. A well-run property needs both answers, and in a commercial context they are usually the same document viewed from two directions.

What goes into a component inventory?

The inventory is the foundation, and it is where a plan is most easily, and quietly, wrong. CAI’s current standards (2023) use a three-part test: the association or owner is obliged to maintain or replace the item, the need and timing can be reasonably anticipated, and the total cost is material and can be reasonably estimated. CAI explains material as more than the operating budget can readily absorb, so that funding it from operations would distort the year it lands.

That last test is what keeps a plan honest in both directions. Ceiling tiles fail the significance test and belong in operating. A roof passes all three. A chiller passes. Interior corridor finishes usually pass on cost even though they are cosmetic. And an item that is genuinely unpredictable — foundation movement, latent envelope defects — does not belong in the reserve schedule as a line item, because putting a made-up date on an unknowable event makes the whole document less credible, not more.

A serviceable commercial inventory usually spans:

• Envelope — roof systems by section and type, facade and sealants, windows and curtain wall, waterproofing.
• Mechanical — chillers, boilers, rooftop units, air handlers, pumps, cooling towers, building automation controls.
• Electrical — service equipment and switchgear, distribution, generators, lighting systems where a full relamp or retrofit is a capital event.
• Vertical transportation — elevator modernization, cab interiors, controllers.
• Site — asphalt and concrete, parking decks and their structural repairs, retaining walls, site lighting, landscaping hardscape.
• Life safety — fire alarm panel replacement and sprinkler system components at end of life.
• Interiors — common-area finishes, corridors, lobbies, amenity spaces.

Two inventory habits pay for themselves. First, record quantity and unit, not just the item — “roof” is not a component, but an illustrative entry such as “38,000 sq ft TPO, north section, installed 2014” is. Second, capture the install or last-replacement date at the same time, because remaining useful life estimated from a real install date is a forecast, while remaining useful life estimated from a guess is a wish.

How do you estimate remaining useful life honestly?

Published useful-life tables are the starting point, not the answer. HUD’s Estimated Useful Life table (opens in a new tab) for multifamily capital needs assessments, for example, assigns one standard life to each component type: 15 years for a TPO or EPDM low-slope roof or a packaged rooftop HVAC unit, and 25 years for a gas-fired sectional boiler or asphalt pavement (as of October 2026). Those values are useful for a component you have never seen, and HUD lets the assessor set a different remaining life when the evidence justifies it.

For a component you can see, condition beats the table. A rooftop unit that has been on quarterly preventive maintenance for a decade and a chemically-treated water loop is a different asset from an identical unit that has been run to failure and patched, and no age-based table distinguishes them. CAI’s standards make the same point: a component’s useful life should reflect the preventive maintenance actually performed. This is a major accuracy gain in capital planning, and it is why the plan should be built from the maintenance record rather than beside it — the work-order history for an asset is the condition assessment, accumulated for free over years. Buildings running a real preventive maintenance program with asset-level history can date their components from evidence; buildings without one are estimating from install dates and hope.

Two adjustments belong in every honest estimate. Escalation: a replacement fifteen years out does not cost today’s price, and a plan that ignores construction cost escalation will underfund every long-dated item in it. Interdependency: components fail in clusters and get replaced in clusters. A roof replacement is the right time to move the rooftop units that sit on it; a garage repair is the right time to redo the lighting and the striping. Sequencing related work saves real money, and a year-by-year table that ignores adjacency schedules three mobilizations where one would do.

How much should a building fund, and how?

CAI’s standards describe three funding goals. Full funding aims to keep the reserve balance at or near the fully funded balance: each component’s current replacement cost times the share of its useful life already used, summed across all components. Threshold funding targets keeping the balance above a chosen floor, in dollars or in percent funded. Baseline funding targets never running out of cash — the balance approaches zero but does not go negative. CAI calls baseline the riskiest goal and does not recommend it as a long-term plan.

The comparison figure most boards learn first is percent funded: the actual reserve balance divided by the fully funded balance. It is a useful health indicator and a badly abused one; CAI’s standards say it is not by itself a measure of adequacy. A low percent funded is a risk signal, not a verdict; what actually matters is whether the funding plan covers the expenditures in the years they land, without a special assessment and without deferring work into failure. For an illustrative example, a building at forty percent funded with a credible plan and a light near-term schedule can be in better shape than one at eighty percent with a chiller, a roof, and an elevator all landing in the same three years.

Commercial owners outside the association world have more room to choose: they can fund from operations, from a replacement-reserve escrow required by a lender, or from financing timed to the work. Lender reserves can be sized from that forecast: Fannie Mae’s multifamily guide (opens in a new tab), for example, sizes a loan’s replacement reserve from the PCA’s cost estimates, out to the sooner of two years past maturity or 12 years after origination (as of October 2026). What owners cannot do is skip the forecast — a lender’s reserve requirement and a buyer’s diligence both ask the same question, an insurer can ask it too, and “we handle it as it comes” is an answer that prices into the deal against you.

Two failure modes to name. Deferral is a loan at an unfavorable rate: pushing a roof three years past its life risks three years of interior water damage, tenant disruption, and emergency repair, and the total can exceed the cost of the replacement it postponed. And a plan that is never revised is not conservative, it is just old — a capital plan should be updated annually against actual spend and condition, with a fresh site-based review of the physical analysis every three to five years. CAI’s standards call a site-visit update at least every third year best practice, and Virginia and Maryland law require a new or updated study at least every five years for the associations they cover.

What is different about associations in Virginia, Maryland, and DC?

For community associations, reserve planning is not only prudent, it is regulated — and the DMV’s three jurisdictions do not regulate it identically. As of October 2026, Virginia’s Condominium Act (Va. Code § 55.1-1965) (opens in a new tab) and its Property Owners’ Association Act (§ 55.1-1826) require a reserve study at least once every five years and a review of its results at least annually; the condominium rule applies unless the condominium instruments provide otherwise. Maryland’s Real Property § 11-109.4 (opens in a new tab) (residential condominiums) and § 11B-112.3 (homeowners associations whose reserve components total at least $10,000) require an independent study, an update at least every five years, a funding plan, and a review of the study with each annual budget; those rules were phased in by county between 2020 and 2023. The District’s Condominium Act differs again: it gives the association the power to budget for reserves (D.C. Code § 42-1903.08) and requires the resale certificate to state the status and amount of reserves (§ 42-1904.11), but it does not mandate a reserve study.

Because these statutes have been amended in recent years and the details differ by association type, confirm the current requirement for your association type in your jurisdiction with association counsel before you rely on an interval you remember. A board planning to the strictest of the three is unlikely to be caught out by an amendment.

A single owner of a commercial building faces no equivalent statute, but they face the functional equivalent through their lender’s reserve requirements and their buyers’ diligence, which is why the discipline converges even where the law does not. Boards navigating this alongside everything else on their plate may find our guide to what boards should expect from HOA management a useful companion.

How does the plan connect to the operating budget?

Often badly — and the disconnection is expensive in both directions.

Capital and operating are separate budgets but a single physical reality. Deferring preventive maintenance shortens asset life and pulls capital expenditures forward; funding maintenance properly pushes them back. When those two budgets are managed by different people looking at different documents, the operating side takes credit for a saving that the capital side pays for later, and nobody connects the two events.

The mechanism that connects them is asset-level data. When work orders, costs, downtime, and PM history are recorded against the specific asset — this chiller, not “HVAC” — the operating record becomes the evidence base for the capital forecast, and the repair-versus-replace decision stops being an argument and becomes arithmetic. That is a large part of what a CMMS is for, and it is why Levaru builds capital plans out of the maintenance system rather than out of a spreadsheet that references it. It also gives the plan the one thing spreadsheets never have: a defensible answer to “how do you know?” The same logic underlies benchmarking the operating side, which we cover in facility management cost per square foot.

What makes a capital plan fail?

Five failure modes are common. An inventory built from a walkthrough and a template rather than from real quantities and install dates. Age-based life estimates never reconciled to condition, so a well-maintained asset and a neglected one carry the same replacement year. No escalation, which systematically underfunds every long-dated item. No sequencing, which schedules three separate mobilizations onto the same roof within a few years. And the simplest: no annual update, so the plan describes a building that existed when it was written and is quietly abandoned the first time reality diverges from it.

Every one of those is a process problem rather than a technical one, which is the encouraging part. A capital plan does not require a forecasting breakthrough. It requires an accurate inventory, condition evidence from the maintenance record, honest escalation, deliberate sequencing, and someone whose job it is to revisit the document every year.

Frequently asked questions

How often should a reserve study or capital plan be updated?

Review and update the numbers annually against actual spending and current conditions, and rebuild the physical analysis — a fresh on-site condition assessment — every three to five years. CAI’s Reserve Study Standards call a site-visit update at least every third year best practice. As of October 2026, Virginia and Maryland require the associations they cover to get a new or updated study at least every five years; treat that legal interval as a minimum, not a target. An annual review is quick when the underlying asset data is current and slow when it is not, which is itself an argument for keeping the asset record live.

How far out should the plan look?

CAI’s Reserve Study Standards require a funding plan covering at least 30 years, a horizon long enough to catch the second replacement of short-lived components and to smooth funding across the big-ticket years. PCA reserve tables are shorter: Fannie Mae’s multifamily guide uses the loan term plus two years, up to 12 years, as of October 2026. The near years should carry real estimates; the far years exist to prevent an underfunded surprise, and everyone should understand that a year-24 number is a planning placeholder, not a bid.

Who should prepare it?

For an association with a statutory requirement, use a provider qualified to meet that requirement — the reserve-study credentials exist precisely so a board can verify competence. Maryland’s statute, for example, accepts a CAI Reserve Specialist, an APRA Professional Reserve Analyst, a Maryland-licensed architect or engineer, or someone who has prepared or helped prepare at least 30 reserve studies in the prior three calendar years. For commercial owners, the useful qualification is different: someone who can read the building’s own maintenance data, not only walk it. Whoever prepares it, the deliverable should include the component inventory in a form you can maintain, not just a PDF summary — a plan you cannot update is a plan with a one-year shelf life.

What is the difference between a capital expenditure and a repair?

Practically, a repair restores a component to service without extending its life materially, and a capital expenditure replaces or substantially extends it. The line matters for accounting and tax treatment. For tax purposes, the IRS tangible property regulations treat a cost as an improvement if it is a betterment, a restoration, or an adaptation to a new use, and the definitive answer for a specific item belongs to your accountant. For planning purposes, the more useful test is the budget-distortion one: if funding it from the operating budget would visibly damage that year, plan for it as capital regardless of how it is later booked.

Is a reserve study required for a commercial property?

No statute requires one the way association law can, but the functional requirement arrives through other doors: lenders can require funded replacement reserves (Fannie Mae’s multifamily guide requires them on some loans), buyers commission a property condition assessment during diligence, and insurers can ask about building condition at underwriting. An owner without a plan is not exempt from the question — they simply answer it later, with less control over the framing.

Can a capital plan reduce spending, or does it just schedule it?

Both, in ways that are measurable. Sequencing adjacent work eliminates duplicate mobilizations. Competitive procurement is available for planned work and not for emergency work. Planned replacements can be timed to tenant turnover and to seasons that price better. And catching a component before failure avoids the collateral damage — the water intrusion, the tenant disruption, the emergency premium — that makes a run-to-failure roof cost more than a replaced one. The plan does not make the chiller cheaper; it removes the surcharge you pay for being surprised.

A capital plan is not a forecast of the future so much as a refusal to be surprised by it. If your building’s next ten years of major spending currently live in a few people’s heads and a folder of vendor proposals, that is the gap worth closing — and it is what our capital planning work is built to close for owners and boards across the DMV.