Asset Classification Evidence and Documentation
Publication 5653 lists 13 principal elements of a quality cost segregation study and nine report elements. Classification must be substantiated with records, a site inspection, and a written legal analysis. Rule-of-thumb percentages without documents are the weakest approach in that guide. Educational information, not tax advice.
Educational information, not tax advice.
Updated 2026-10-14
Key takeaways
- The IRS has not established required qualifications or a required report format; Publication 5653 describes what examiners look for and states it is not an official pronouncement of law.
- The 13 quality-study elements include expertise, methodology, documentation, interviews, common nomenclature, legal analysis, take-offs, reconciliation, and listing of section 1245 property.
- Detailed engineering from actual cost records is described as generally the most accurate approach; residual estimation and rules of thumb are described as less reliable.
- Boddie-Noell rejected after-the-fact percentage estimates; AmeriSouth rejected most classifications that lacked item-by-item support; Peco Foods bound the buyer to a written section 1060 allocation.
- Functional allocation is an examiner-accepted method only for primary and secondary electrical distribution, not for plumbing or HVAC.
Classification opinions without evidence are just opinions. Treas. Reg. §1.48-1 still supplies the legal boxes: tangible personal property except land, buildings, and structural components; a building as a roofed enclosure used for shelter, housing, or working, office, parking, display, or sales space; and a structural-component list that includes walls, partitions, floors, ceilings, windows, doors, central HVAC, plumbing fixtures, electric wiring, lighting fixtures, sprinkler systems, and fire escapes. Hospital Corp. of America held that those tests apply under MACRS. Whiteco’s six permanence factors are a court test, not a statute. None of those authorities is a substitute for a file an examiner can test.
Publication 5653 Chapter 3 lists six study approaches: detailed engineering from actual cost records; detailed engineering cost estimates; contractor surveys; residual estimation; sampling or modeling; and rules of thumb. The guide calls the actual-cost engineering approach generally the most methodical and accurate, and generally applicable to new construction. Residual estimation costs only short-lived assets and dumps the remainder into the building; the guide says it is simpler and can be less accurate, and that mixed techniques can skew results toward section 1245. Rules of thumb apply industry-average percentages with little documentation; examiners should view them with caution and should not categorically reject them. The Service has not established required standards for preparing studies. Substantiation using actual costs is more accurate than using estimates. Applying a pro-rata step-up factor to estimates to force a purchase-price reconciliation is not an acceptable method.
Chapter 4 lists 13 principal elements of a quality study: preparation by someone with expertise; a detailed methodology description; appropriate documentation (drawings, pay applications, site visit with photographs; for acquisitions, land valued first and studies on used property “should be performed by qualified appraisers,” citing AmeriSouth); interviews; common nomenclature; a standard numbering system such as CSI MasterFormat; a legal analysis; unit costs and engineering take-off; asset grouping tied to the ledger; reconciliation of allocated costs to actual costs; treatment of indirect costs; identification of section 1245 property; and related issues such as section 263A and accounting-method change. Nine report elements follow, from a summary letter through exhibits. The guide is examiner guidance, not law, and it is not an “IRS-approved study” certificate.
Court files show what happens when the evidence is thin. Boddie-Noell found restaurant percentage estimates far too unreliable when they rested on after-the-fact studies of seven units extrapolated to the rest, without contemporaneous records. AmeriSouth kept most of a thousand-plus component study with apartment buildings because the items were structural components of those buildings. Peco Foods, Inc. v. Commissioner held that a written section 1060 purchase-price allocation bound the buyer, so a later study could not re-subdivide “buildings” inconsistently with the agreed schedules. Scott Paper and Morrison show the opposite quality: a load-based electrical split tied to design. A.C. Monk rejected allocation on different facts. Courts split on the same asset types — partitions, HVAC, sprinklers, electrical — depending on facts and records.
CostSeg SI has described a proposed Super Intelligence workflow covering extraction, normalization, classification, estimation, quality review, and engineer certification. Those product steps are proposed. They are not a claim that drawing extraction, published unit-cost application, or any accuracy or time-saving figure is operational. Publication 5653 names R.S. Means alongside Marshall Valuation Service as an example of a reliable published source of costing data that examiners have seen, and Chapter 5 shows that such estimates can diverge from actual contractor unit costs. Naming a published source is not an endorsement and is not a CostSeg SI product feature.
The minimum file for classification
A closing statement or schedule of values, drawings, inspection photographs, takeoff or invoice mapping, a citation to the regulation or case for each class, and a tie-out to depreciable basis are the documents a CPA and, if needed, an examiner can test. For electrical claims, add one-line diagrams and demand-load calculations. For dedicated circuits, add panel directories. For coverings and millwork, add attachment photographs. For site work, add a land-versus-improvement split.
What examiners are told about electrical, plumbing, and HVAC
Chapter 8.A discusses functional allocation of primary and secondary electrical distribution by design or demand load. A properly applied allocation of that kind should not be challenged, according to the guide. The guide does not extend that method to plumbing or HVAC as an IRS-accepted allocation. Restaurant matrices say allocation of HVAC is not appropriate. Plumbing fixtures are listed structural components. Dedicated equipment hook-ups are identified, not percentage-spread.
Nomenclature and double counting
Chapter 4 warns against creative labels, such as calling sewage piping “process piping.” It also warns against double-counting separately acquired furniture and equipment. Totals must reconcile. Residual estimation that never reconciles is a quality flag in Chapter 3.
Acquisitions and method changes
For purchased property, land is valued first at highest and best use. Replacement-cost estimates should be adjusted for physical deterioration and functional obsolescence. A written section 1060 allocation can limit later reclassification, as Peco Foods held. Reclassifying via a study is generally a change in accounting method requiring Commissioner consent, which the guide notes in Chapter 4.C.13. That is a CPA filing issue, not a website conclusion.
How to read this library
Each asset page in this library is educational information, not tax advice. Industry matrices in Publication 5653 are examiner guidance for general fact patterns in named industries. Court outcomes are examples. An item is not five-year or 15-year property as a rule because another taxpayer prevailed on different facts. Qualified review belongs with the taxpayer’s own adviser before any return position is taken.
Sources
- 1IRS Publication 5653 — Cost Segregation Audit Techniques Guide (rev. 2-2025)(opens in a new tab)
Written for IRS examiners. The guide states it is not an official pronouncement of law and cannot be cited as authority. It predates the July 2025 law.
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10