top of page
Search

Replacement Reserves in Commercial Real Estate: What They Are, Why Lenders Model Them, and What to Assume by Asset Class

  • 2 days ago
  • 19 min read

By Michal Mohelsky, J.D. — Principal, MMCG Invest, LLC | July 2026



In most underwriting models, the replacement reserve occupies a single row. It sits between the operating expenses and the net operating income, it rarely exceeds two or three percent of revenue, and almost nobody argues about it. Two hundred fifty dollars a unit. Fifteen cents a square foot. Four percent of hotel revenue. The figures are recited in credit memos the way liturgy is recited in church — from memory, without much curiosity about where they came from.


We got curious. What we found is that the industry's standard reserve conventions are historical artifacts, most of them set two decades or more ago and never indexed to anything. The $250-per-unit multifamily floor entered HUD practice through Mortgagee Letter 2010-21, which raised a prior $150 figure (1). The CMBS per-square-foot floors — $0.15 for retail, industrial, and self-storage, $0.20 for office, $50 to $100 per manufactured-housing pad, 4 to 5 percent of revenue for hotels — trace to conduit originator underwriting conventions disclosed in securitization prospectuses filed with the SEC in 2003 and 2004 (2). Since those documents were filed, the dollar has lost well over 40 percent of its purchasing power (3), and the Engineering News-Record Construction Cost Index has risen roughly 23 to 24 percent since 2020 alone (4). The floors barely moved.


This matters because the reserve is not a formality. It is the point where engineering, valuation, and credit meet. Get it wrong and the appraised value is wrong, the debt service coverage ratio is wrong, the loan is sized wrong, and — five or fifteen years later — the roof still needs replacing whether the model acknowledged it or not. This article covers what a replacement reserve actually is, why lenders and appraisers model it, where the conventional numbers came from and what the evidence says they should be, and how we build a defensible figure in a lender-grade feasibility study.


Line chart, "Frozen Floors vs. Construction Costs, 2003–2026." Two series indexed to 100 in 2003: (a) the $0.15/SF CMBS retail reserve floor, a flat line; (b) ENR Construction Cost Index annual values (2003 base through March 2026 = 14,156.75).


What a replacement reserve is — and what it is not

A replacement reserve (also called a reserve for replacement or a replacement allowance) is an annualized allowance for the periodic replacement of short-lived building components: the roof membrane, the rooftop HVAC units, the parking lot, the water heaters, the appliances, the elevator machinery, the hotel's furniture and case goods. These components wear out on cycles of seven to twenty-five years — much faster than the structure itself — and the reserve smooths their lumpy replacement cost into a level annual figure the income statement can carry (5).


It helps to distinguish the reserve from its four neighbors, because credit files routinely conflate them. Repairs and maintenance is an operating expense: the recurring service, cleaning, and minor fixes that keep components alive. Capital expenditure is the actual cash spent when a component is replaced; the reserve is a forecast, capex is a fact, and over a long hold the two should converge. Tenant improvements and leasing commissions are re-leasing costs specific to commercial space and are modeled separately from structural reserves. Deferred maintenance is an existing deficiency — a cost of today, not a provision for tomorrow — and property condition assessors price it as an immediate repair, typically escrowed at 100 to 125 percent of the estimate at closing (6).


One more distinction, and it surprises people: the reserve line is neither a tax concept nor an accounting one. A deposit into a funded reserve is generally not deductible when made; the deduction arrives only when the money is actually spent, either immediately as a repair or over time through depreciation of the capitalized replacement (7). Under GAAP, a discretionary reserve is not even a liability. The reserve is purely an underwriting construct — which is exactly why its treatment is a matter of convention, and why conventions this stale deserve scrutiny.


Above the line, below the line, and the consistency rule

Appraisal practice tolerates two treatments. "Above the line" deducts the reserve before net operating income, reducing NOI and, at a given capitalization rate, reducing value. "Below the line" leaves NOI untouched and shows the reserve as a deduction in the cash-flow section. The building does not care where the analyst books the allowance — the same roofs will fail on the same schedule — but the reported NOI changes, and so does everything derived from it.


The trap is inconsistency. Capitalization rates are extracted from comparable sales by dividing each sale's NOI into its price. If those comparable NOIs did not deduct reserves, applying the resulting cap rate to a subject NOI that does deduct reserves systematically understates the subject's value — by the reserve amount divided by the cap rate. The arithmetic is not subtle: a $50,000 annual reserve at an 8 percent cap rate is $625,000 of value (8). We have seen appraisals move by that much on nothing more than an unexamined line placement. Institutional convention (NCREIF, the Mortgage Bankers Association) holds reserves below the NOI line; HUD and the agencies underwrite above it. Both are defensible. Mixing them is not, and the Minnesota Supreme Court's decision in Carson Pirie Scott & Co. v. County of Hennepin — affirming a $0.20-per-square-foot reserve deduction in an income-approach tax valuation — is a useful reminder that the treatment gets litigated (9).


Why lenders insist on the deduction

Strip away the mechanics and the reserve does three jobs in a credit file.


It disciplines the valuation. Direct capitalization assumes a single stabilized year repeats forever. Lumpy capital events cannot live inside a single year, so the smoothed allowance stands in for them. An NOI with no reserve is a fiction — a building that never ages — and capitalizing a fiction produces a fictional value. Discounted cash flow can do better by scheduling the actual outlays (the roof in year eight, the mill-and-overlay in year twelve), which is why institutional DCF models often replace the level allowance with an explicit capital schedule.


It sizes the loan. Consider a property producing $1,000,000 of NOI before reserves, underwritten to a 1.25x coverage ratio at a 7 percent mortgage constant. With no reserve, supportable debt service is $800,000 and the loan is roughly $11.4 million. Impose a $100,000 reserve and NOI falls to $900,000, supportable debt service to $720,000, and the maximum loan to about $10.3 million — a 10 percent haircut. The same $100,000 simultaneously compresses the income-approach value from $12.5 million to $11.25 million at an 8 percent cap rate, tightening the loan-to-value constraint in lockstep. One assumption, two binding constraints (10).


The government-guaranteed programs our clients use make this structural rather than optional. USDA's OneRD regulation defines the debt service coverage ratio itself as EBITDA "less reasonably expected replacement capital expenditures" divided by annual debt service — the reserve lives inside the ratio by regulatory definition (11). SBA SOP 50 10 8, effective June 1, 2025, sets coverage floors of 1.15x for 7(a) loans over $350,000 and, effective March 1, 2026, 1.10x for smaller 7(a) loans, with the 504 program at 1.0x (12). No SBA rule mandates a funded reserve — but a feasibility study that models a realistic one compresses projected coverage toward those floors, and a study that omits it invites rework at credit committee.


Grouped bar chart, "One Assumption, Two Constraints." X-axis: annual reserve assumption ($0 / $50K / $100K / $150K). Two series: maximum supportable loan (1.25x DSCR, 7% constant) and capitalized value (8% cap) on $1.0M pre-reserve NOI.


It corrects for a documented bias. The third job is the least discussed and the most important: the standard reserves demonstrably understate what buildings actually consume. Green Street built its "economic cap rate" framework on precisely this point, observing that the standardized reserves used by market participants are far smaller than the capital expenditure historically incurred by public and private owners (13). The hotel sector supplies the cleanest evidence because contractual reserves and actual spending are both disclosed: the ISHC/HAMA CapEx study, first published in 1995 and updated through its sixth edition in 2023, found capital spending reached 9 percent of revenue and a record $5,147 per available room — roughly double the traditional 4 percent FF&E reserve — with repairs and maintenance consuming a further 4.4 percent of revenue on top (14). The first study concluded thirty years ago that the convention was inadequate. The convention has moved from roughly 3 percent to 4 or 5. The gap persists.


The academic literature explains why the gap is a value problem, not a cash-flow inconvenience. Bokhari and Geltner, working from more than 100,000 transactions, measured real net depreciation of U.S. income property at about 1.5 percent of total property value per year — 1.8 percent for new buildings, declining with age — and, adding back the capital expenditure owners actually made, gross capital consumption of roughly 3.5 percent of property value annually for commercial assets and 4.4 percent for apartments (15). UK research funded by the Investment Property Forum found the mirror image: properties receiving higher capital expenditure exhibited measurably lower rental depreciation (16). Reserves offset physical wear. They cannot buy back functional or economic obsolescence — nothing can — which is why an adequate reserve is necessary but not sufficient to hold value, and an inadequate one is a slow leak in the collateral.


Where the number actually comes from

A defensible reserve is engineered, not recited. The instrument is the property condition assessment governed by ASTM E2018-15, which inventories major components, assigns each an expected useful life, observes its effective age, and derives a remaining useful life (17). Components are scheduled for replacement in the year their remaining life expires; costs are stated in current dollars and then escalated; the resulting year-by-year requirement is leveled into an annual per-unit or per-square-foot deposit. Horizons vary by program — commercial PCAs commonly run twelve years matched to a loan term, HUD's CNA e-Tool runs twenty.


The leveling math matters more than most underwriters realize. The reserve-study profession — which, unlike commercial practice, has codified standards through the Community Associations Institute — recognizes distinct funding methods: straight-line averaging, a sinking fund that credits interest earnings on accumulating balances, and a cash-flow method that solves for the minimum level deposit keeping the balance above zero across the horizon (18). Commercial real estate almost universally uses the crudest of the three, a flat straight-line figure that ignores both interest earnings and the clustering of expenditures in particular years. The condominium world shows where lax funding discipline leads: the April 2026 industry analysis of more than 100,000 reserve studies found 74 percent of associations funded below the 70 percent-of-need threshold considered strong — the highest underfunding rate ever recorded — and post-Surfside Florida now prohibits waiving reserves for structural components entirely (19).


Two adjustments separate a lender-grade schedule from a generic one, and both are supported by formal standards that commercial practice rarely cites. The first is climate. Component life is not a national constant: a U.S. Army Corps of Engineers study documented coastal salt air cutting air-conditioning equipment life by as much as half; peer-reviewed pavement research confirms freeze-thaw cycling and de-icing salt materially shorten asphalt fatigue life; hail-alley and high-UV markets consume roofs faster than their nameplate ratings suggest (20). ISO 15686 — the international service-life planning standard — formalizes exactly this adjustment through its factor method, converting a reference service life into an estimated one based on environment, use, and maintenance level (21). The second is cost. National-average replacement costs must be localized: the RSMeans City Cost Index runs 40 to 60 percent above the national average in the most expensive coastal metros and 15 to 25 percent below it in parts of the rural Midwest (22). A schedule that adjusts neither axis is a schedule for a building that exists nowhere.


Horizontal bar chart, "Same Component, Different Life." Expected useful life ranges for a low-slope membrane roof, packaged rooftop HVAC, and asphalt paving under three exposure profiles (temperate baseline; hail/freeze-thaw belt; coastal salt/high-UV).


New buildings, old conventions

Because a large share of our practice is ground-up development feasibility, we spend a lot of time on a question the conventions answer badly: should a brand-new building reserve at the same rate as a stabilized twenty-year-old asset?


The engineering answer is no. Every component starts with full remaining life, and the early years are blanketed by warranties — one to two years on systems workmanship, five to ten on HVAC parts and compressors, ten to thirty on commercial roofing membranes, up to ten on structure (23). The vintage curve in the hotel data is unambiguous: properties under five years old spent under 2.5 percent of revenue on capital projects, while hotels past fifteen years spent more than 8 percent, with spending peaking as the major systems turn over in years sixteen through twenty-five (14). Capital need is back-loaded. A graduated reserve — minimal in years one through five, escalating as warranties expire, reaching the stabilized figure by the first replacement cycle — matches the physics.


The credit answer is more complicated. HUD's Section 221(d)(4) program requires new construction to fund reserves from day one at the greater of 0.60 percent of total cost (0.40 percent of loan amount for substantial rehabilitation) or $250 per unit (24), and lenders generally prefer level funding for the simple reason that it builds a cushion before discipline can lapse. Our practice in a ground-up pro forma is to show both: the engineering-optimal graduated schedule in the lease-up years, the mandated deposit where a program requires one, and — critically — a full stabilized reserve in the stabilized year used for the exit valuation. Capitalizing a warranty-shielded year-two NOI into a terminal value is one of the quieter ways a development model overstates itself. The rest of the construction-phase reserve stack — interest reserve, hard-cost contingency (now up to 15 percent of the construction budget under SBA 504 rules effective September 30, 2025), soft-cost contingency, operating deficit and working capital escrows — is a separate architecture with its own logic, and none of it substitutes for the recurring reserve that begins at conversion (25).


What each loan program actually requires

The most useful organizing distinction in this entire subject is between reserves that are escrowed — funded monthly into a lender-controlled account with documented release procedures — and reserves that are merely modeled as an expense line in the projections. Confusing the two produces both kinds of error: borrowers blindsided by an escrow they did not budget, and feasibility studies that treat the absence of an escrow as the absence of the cost.

Program

Requirement

Escrowed or modeled

HUD 223(f)

Minimum $250/unit/year, sized by PCNA, plus initial deposit (1)(24)

Escrowed; releases via Form HUD-9250, three bids above the larger of $50,000 or 20% of balance (26)

HUD 221(d)(4)

Greater of 0.60% of cost (new) / 0.40% of loan (rehab) or $250/unit (24)

Escrowed from day one

Fannie Mae / Freddie Mac

Greater of PCA-derived schedule or floor (working convention $250–$300/unit); NOI underwritten net of reserves (27)

Escrowed for higher-risk tiers; may be underwritten-only for the strongest deals

CMBS conduit

Originator floors: $0.15/SF retail, industrial, self-storage; $0.20–$0.25/SF office; $250/unit multifamily; $50–$100/pad MHC; 4–5% of revenue hotels — applied "notwithstanding actual escrows," with separate TI/LC reserves (2)

Underwriting deduction; escrows deal-specific

SBA 7(a) / 504

No mandated reserve; realistic reserve belongs in the feasibility projections and presses against DSCR floors of 1.15x / 1.10x / 1.0x (12)

Modeled

USDA B&I (7 CFR 5001)

No fixed reserve, but the DSCR definition nets replacement capex; escrows set deal-by-deal in the letter of conditions (11)

Modeled, inside the ratio

USDA Community Facilities / Water-Waste (7 CFR 1780)

Debt service reserve equal to one average annual installment, accumulating at one-tenth per year, plus a short-lived asset reserve scheduled in the preliminary engineering report (28)

Escrowed by regulation — the strongest codified reserve in the rural portfolio

Matrix graphic, "Escrowed vs. Modeled." Programs on one axis, reserve treatment (escrowed account / underwriting deduction / inside the DSCR definition). Source: HUD, FHFA program guides, SBA SOP 50 10 8, 7 CFR 5001 & 1780; MMCG.


Two SBA-specific points deserve emphasis because they define our lane. First, most SBA collateral is owner-occupied — 51 percent occupancy for existing buildings, 60 percent for new construction — which collapses the landlord-tenant division of capital responsibility. The reserve becomes, functionally, a business capital budget, and it must be captured in the global cash flow analysis or it is captured nowhere. Second, for special-purpose properties operated as going concerns — hotels, car washes, gas stations, self-storage — SOP 50 10 8 requires a going-concern appraisal allocating separate values to land, building, equipment, and intangibles (12). That allocation forces the analyst to keep the real-property reserve (roof, HVAC, paving) distinct from the equipment reserve (tunnel machinery, kitchen lines, FF&E). Modeling both in full across overlapping components double-counts the capital need; modeling only one understates it.


Benchmarks by asset class — floors versus evidence

The table below is the reference we are asked for most often. Read the two right-hand columns together: the first is what the lending conventions assume, the second is what the evidence and practitioner experience suggest a realistic figure looks like. The floors are floors. They are not estimates.

Asset class

Convention / floor

What the evidence supports

Multifamily

$250/unit/yr (HUD, CMBS); $250–$300 (agency practice) (1)(2)(27)

$250–$400+/unit by vintage; realized apartment capital consumption runs multiples of the floor (15)

Office

$0.20–$0.25/SF, TI/LC separate (2)

$0.15–$0.25/SF Class A structural, higher for older stock; rollover timing is the real risk

Retail

$0.15/SF floor (2)

$0.20–$0.35/SF in current practice

Industrial / small-bay flex

$0.05–$0.15/SF; NNN shifts most capex to tenants (2)

Read the lease — roof and structure carve-outs put six-figure replacements back on the landlord

Self-storage

$0.12–$0.15/SF (2)

Same range; note the standing tension between appraisals that exclude reserves and operators who fund them

Hotels (limited/select service)

4% of revenue, ramping 1–2% in early years (2)(14)

Realized spending 8–9% of revenue; brand PIP obligations of $5,000–$25,000 per key (mid-market renovations now $35,000–$40,000/key) sit on top of the reserve (14)(29)

Manufactured housing communities

$50–$100/pad (2)

The pad figure ignores the real exposure: private wells, package treatment plants, electrical pedestals, and roads — infrastructure a building-focused PCA routinely under-scopes (30)

RV resorts / campgrounds

~$75/site (survey convention) (31)

Plus utility pedestals, bathhouses, pools — reserve by component, not per site alone

Senior housing / assisted living

$300–$400/unit; industry survey range $181–$525, median ≈ $326 (32)

Equipment- and regulation-intensive; skilled nursing higher still

Fitness

—

The clearest published trade benchmark anywhere: roughly 4% of revenue or $4/SF annually for maintenance capital, plus about 10% of revenue every five years for market-driven refresh (33)

Car wash, QSR, fuel/c-store

—

No per-SF convention is meaningful; reserve from equipment lives (tunnel and wash equipment largely 7–25 years; fryers 7–10; walk-ins 15–20; dispensers and USTs on compliance cycles). Operators reserving 1–2% of sales cannot fund a mid-term franchise remodel (34)

Horizontal bar chart, "Convention vs. Evidence." Paired bars per asset class showing the underwriting floor against the evidence-supported range (hotel shown as % of revenue on a secondary axis or split panel). Chart.js, dark green floors, copper evidence bars. Source: SEC EDGAR conduit filings; ISHC CapEx 2023; ASHA; IHRSA; MMCG.


The pattern across the table is consistent: wherever an asset class produces audited evidence of actual capital spending — hotels through USALI disclosure, apartments through transaction-based depreciation research, fitness through trade-association benchmarking — the evidence lands well above the underwriting convention. Where no evidence exists, the convention is usually just old.


The erosion problem: 2020–2026

Even an analyst who accepts the conventions must confront what the last six years did to them. The ENR Construction Cost Index rose from an annual average near 11,466 in 2020 to 14,156.75 by March 2026 — roughly 23 to 24 percent — before any component-specific effects (4). The components that dominate reserve schedules did worse. The federal phase-down of R-410A refrigerant ended manufacture of new R-410A split systems on January 1, 2025; replacement A2L equipment carried initial wholesale premiums in the 10 to 20 percent range, moderating toward the high single digits by early 2026, while the legacy refrigerant needed to keep old units alive runs three to four times its former price per pound (35). Hotel FF&E product costs rose 15 to 20 percent over five years, with general-contractor costs on renovations up 30 to 40 percent (36).


The insurance market then converted deferred roof capital into an operating-expense problem. Carriers now routinely limit, exclude, or non-renew coverage on roofs past twenty years — ten to fifteen for flat commercial systems — and shift settlement from replacement cost to actual cash value as roofs age, precisely when the payout matters (37). The Minneapolis Fed's multifamily survey recorded premium increases of 14, 22, and 45 percent in consecutive years through 2024 (38). A property that "saved money" by skipping the roof reserve now pays for it through the insurance line, at a worse exchange rate.


Put plainly: a reserve schedule costed on pre-2021 assumptions is understated by roughly a quarter on general components and by more on HVAC and FF&E. Escalating a stale schedule is not a fix. Re-costing it in current dollars is.


The compelled-capital blind spot

Traditional reserve schedules answer one question — when will this component wear out? — and modern regulation increasingly asks a different one: when must it be replaced regardless? Building performance standards now operate in sixteen jurisdictions; New York's Local Law 97 fines $268 per metric ton of CO2e over a building's cap, and roughly 57 percent of covered properties currently exceed their 2030 limits (39). Seismic ordinances in Los Angeles and a dozen other West Coast jurisdictions compel retrofits on statutory deadlines — the typical Los Angeles soft-story retrofit ran about $11,000 per housing unit across the program (40). Retroactive elevator and sprinkler codes, facade inspection ordinances, the EPA's 2024 lead service line replacement mandate, and franchise property-improvement plans all share the same character: deadline-driven capital that a wear-out schedule never sees. A gas boiler with fifteen years of remaining useful life is "fine" on a PCA and a stranded asset under an emissions cap. Our rule in a feasibility study is to reserve for the earlier of wear-out or mandate, and to size any obligation whose deadline falls inside the loan term as a discrete, dated line — not a contingency footnote.


One counterweight worth knowing: capital-funding alternatives have deepened. C-PACE assessments — now active in more than thirty states, with roughly $10 billion of cumulative investment through 2024 — can fund exactly the roof, HVAC, and envelope work a reserve exists for, on twenty- to thirty-year terms matched to component life, with the interest deductible where reserve deposits never were (41). Agency supplemental and green loan programs, and HUD's 241(a) supplemental loan for insured properties, do the same inside their lanes. A rigorous study compares funding a reserve against borrowing at the time of need, because the honest answer — after weighing the tax asymmetry, the opportunity cost of trapped cash, and the very real risk that financing is unavailable at the exact moment a distressed property needs it — differs by deal.


How we build a reserve that survives scrutiny

Our method, in the order we execute it. First, we build the component schedule from the property condition assessment, organized under the UNIFORMAT II element classification so the inventory is complete and auditable rather than ad hoc (42). Second, we adjust on two axes: component lives for the site's actual climate and exposure under the ISO 15686 factor method, and unit costs for the market through the current city cost index — in 2026 dollars, not escalated relics (21)(22). Third, we run the funding math as a sinking-fund or cash-flow calculation with disclosed escalation and interest assumptions, benchmarked against published federal discount rates, instead of a flat per-unit plug (43). Fourth, we overlay the compelled-capital screen — performance standards, seismic and life-safety deadlines, insurance-driven roof exposure, franchise obligations — and reserve to the earlier of wear-out or mandate. Fifth, we present the dual-constraint sensitivity: how the reserve assumption moves DSCR, maximum loan proceeds, and capitalized value together, so the credit committee sees the whole mechanism rather than one row of it.


And we disclose the gap. Where our bottom-up figure exceeds the applicable program floor — and for older, equipment-heavy, or climate-exposed assets it usually does — the study says so, quantifies it, and states the coverage implication. A reserve is a forecast of the building's future told in dollars. Lenders are entitled to an accurate one, and a borrower is far better served by confronting the number at underwriting than by meeting it, unfunded, in year nine.


MMCG Invest, LLC prepares lender-grade third-party feasibility studies for SBA 7(a), SBA 504, USDA B&I, REAP, and Community Facilities financings across more than thirty asset classes. Reserve schedules in our studies are built, sourced, and stress-tested to the standard described above.


Author: Michal Mohelsky, J.D., Principal, MMCG Invest, LLC


Request a Feasibility Study → Book a Meeting: https://calendar.app.google/EJzWEz3GCqLY2jU86



Michal Mohelsky, J.D. | Principal | mmcginvest.com 

Phone: (628) 225-1125




Disclaimer: This report is provided for informational purposes only and does not constitute investment advice. Data presented herein is derived from proprietary MMCG databases and third-party sources believed to be reliable; however, MMCG Invest makes no representation as to the accuracy or completeness of such information. Figures from third-party industry databases have been independently verified and, where appropriate, adjusted to reflect MMCG's proprietary analytical methodology. Past performance is not indicative of future results.




Sources

  1. U.S. Department of Housing and Urban Development, Mortgagee Letters 2010-21 and 2012-25 (Reserve for Replacement minimums).

  2. GE Commercial Mortgage Corp., Prospectus Supplements (Form 424B5), 2003–2004, U.S. Securities and Exchange Commission EDGAR; COMM 2013-LC6 Free Writing Prospectus (office floor revision).

  3. U.S. Bureau of Labor Statistics, CPI Inflation Calculator.

  4. Engineering News-Record, Cost Index Review, March 2, 2026 (Construction Cost Index, 1913 = 100); ENR quarterly cost reports, 2020–2026.

  5. Appraisal Institute, The Dictionary of Real Estate Appraisal, 7th ed. (2022); The Appraisal of Real Estate, 15th ed.

  6. ASTM International, E2018-15, Standard Guide for Property Condition Assessments (immediate repairs and reserve-table provisions).

  7. Internal Revenue Code § 263(a) and Treasury tangible property regulations (T.D. 9636); One Big Beautiful Bill Act (2025) bonus depreciation and § 179 provisions.

  8. PropertyMetrics, "Understanding Replacement Reserves in Commercial Real Estate."

  9. Carson Pirie Scott & Co. (Ridgedale) v. County of Hennepin, 576 N.W.2d 445 (Minn. 1998).

  10. MMCG Invest worked example; methodology available on request.

  11. 7 C.F.R. § 5001.3 (OneRD Guaranteed Loan Regulation, debt service coverage ratio definition).

  12. U.S. Small Business Administration, SOP 50 10 8 (effective June 1, 2025); SBA Procedural Notices 5000-875701 and 5000-876777 (effective March 1, 2026).

  13. Green Street, economic cap rate methodology and glossary.

  14. International Society of Hospitality Consultants, HAMA, and STR/CoStar, CapEx 2023: A Study of Capital Expenditures in the U.S. Hotel Industry (6th ed.), as reported by Hotel Investment Today; ISHC CapEx study series, 1995–2023.

  15. Bokhari, S., and D. Geltner, "Characteristics of Depreciation in Commercial and Multi-Family Property," Real Estate Economics 46(4), 2018; Bokhari and Geltner, "Commercial Buildings Capital Consumption and the United States National Accounts," Review of Income and Wealth 65(3).

  16. Crosby, N., S. Devaney, and V. Law, "Rental Depreciation and Capital Expenditure in the UK Commercial Real Estate Market, 1993–2009," Journal of Property Research 29(3), 2012 (Investment Property Forum research programme).

  17. ASTM International, E2018-15 (expected useful life and remaining useful life framework).

  18. Community Associations Institute, National Reserve Study Standards (funding goals and percent-funded metric).

  19. Association Reserves, HOA Reserves: Industry Insights Report (April 2026); Florida Statutes §§ 553.899, 718.112(2)(g), as amended by SB 4-D (2022), SB 154 (2023), and HB 913 (2025).

  20. U.S. Army Corps of Engineers, ERDC, corrosion-resistant coatings demonstration study (DTIC ADA619778, 2015); peer-reviewed freeze-thaw asphalt fatigue studies, Construction and Building Materials (2020, 2024).

  21. International Organization for Standardization, ISO 15686-8:2008, Buildings and Constructed Assets — Service-Life Planning — Reference Service Life and Service-Life Estimation (factor method).

  22. Gordian, RSMeans City Cost Index; Marshall & Swift/CoreLogic local cost multipliers.

  23. Manufacturer and builder warranty conventions: structural (up to 10 years), systems workmanship (1–2 years), HVAC parts and compressors (5–10 years), commercial roofing membranes (10–30 years).

  24. U.S. Department of Housing and Urban Development, Multifamily Accelerated Processing (MAP) Guide (Sections 223(f) and 221(d)(4) reserve requirements).

  25. SBA Procedural Notice 5000-872764 (504 construction contingency, effective September 30, 2025); HUD MAP Guide (working capital and initial operating deficit escrows).

  26. HUD Handbook 4350.1, Chapter 4 (Reserve Fund for Replacements administration; Form HUD-9250).

  27. Fannie Mae Multifamily Selling and Servicing Guide (Replacement Reserve); Freddie Mac Multifamily Seller/Servicer Guide.

  28. 7 C.F.R. Part 1780 (§§ 1780.35, 1780.39); USDA RUS Bulletin 1780-2 (Preliminary Engineering Report requirements).

  29. HVS renovation cost commentary; Hunter Hotel Advisors, mid-market PIP cost estimates (2024–2025); IHG Formula Blue program disclosures.

  30. Fannie Mae Multifamily Guide, Manufactured Housing Community provisions (private utilities, community age).

  31. National RV park valuation survey convention (per-site reserve); practitioner sourcing.

  32. American Seniors Housing Association, The State of Seniors Housing (replacement reserve survey range).

  33. IHRSA (Health & Fitness Association), Health Club Business Handbook (maintenance capital and refresh benchmarks).

  34. Equipment manufacturer service-life data (car wash, foodservice, fuel dispensing); Franchise Disclosure Document Items 11 and 17 (remodel obligations).

  35. U.S. EPA, AIM Act Technology Transitions rule (R-410A production restrictions effective January 1, 2025); ACDirect, 2025 A2L White Paper; industry pricing surveys, 2025–2026.

  36. Hospitality Net, hotel FF&E and general-contractor cost inflation commentary (2024–2025).

  37. Insurance Information Institute (wind/hail share of property claims); carrier roof-age underwriting practice surveys, 2024–2026.

  38. Federal Reserve Bank of Minneapolis, multifamily property insurance survey (2024).

  39. N.Y.C. Admin. Code § 28-320 (Local Law 97); Urban Green Council compliance analyses; Institute for Market Transformation, building performance standards tracker (2026).

  40. Porter, K., Los Angeles Soft Story Retrofit Benefits and Costs (2023); City of Los Angeles Ordinance 183893.

  41. PACENation, C-PACE market data through 2024; state enabling legislation tracker.

  42. ASTM International, E1557, Standard Classification for Building Elements and Related Sitework — UNIFORMAT II.

  43. ASTM International, E917, Standard Practice for Measuring Life-Cycle Costs of Buildings and Building Systems; NIST Handbook 135 and Annual Supplement (federal discount rates).

 
 
 
bottom of page