Extraction & Allocation
~10 min read · Back land value out of improved sales by extraction or allocation ratios.
When vacant-land sales don't exist, back the land out of improved sales: extraction subtracts the depreciated improvements; allocation applies a typical land-to-value ratio. Two rough tools, honestly labeled, tested for exactly what they assume.
Extraction
Extraction: land value = improved sale price − depreciated cost of the improvements (cost new less all depreciation, estimated for the comp). Works best where improvements contribute LITTLE — older homes on valuable land, rural properties with modest structures — because errors in the improvement estimate then barely move the residual. On a new-construction sale, extraction is circular (the improvement estimate IS most of the price).
- Land = sale price − depreciated improvement cost
- Strongest when improvements are a small share
- Each comp extraction = one land indication
Allocation
Allocation applies a typical ratio of land value to total value — extracted from markets where both land and improved sales exist — to the subject's market: if comparable neighborhoods show land running 30% of total value, a $500,000 improved property implies $150,000 land. Ratios drift with appreciation cycles (land share climbs in hot markets) and differ by age and class — allocation is a check and a last resort, not a precision tool.
- Land ≈ total value × typical land ratio
- Ratios from paired markets; class- and age-specific
- Corroboration tool, rarely the lead
Where they fit
The land-valuation toolbox ranks: sales comparison (always preferred), then extraction, allocation, and the income-family methods (development, residual, ground-rent capitalization) where their data exists. USPAP's standard is credibility — a Standard 1 land value built solely on a stale ratio invites review trouble; stack methods and reconcile.
Worked example
No vacant lots have sold for years in a built-out beach town. Recent improved sales: a 1955 cottage (dep. improvement cost ≈ $95,000) sold $890,000; a 1962 ranch (dep. cost ≈ $120,000) sold $915,000; a 2021 custom build (dep. cost ≈ $780,000) sold $1,560,000. Inland towns with active lot markets show beach-comparable land ratios near 80%. Value a typical lot.
Extraction on the old-improvement sales: 890,000 − 95,000 = $795,000; 915,000 − 120,000 = $795,000 — two clean indications, reliable precisely because the cottages barely matter to the price (a 20% error in their depreciated cost moves the residual ~2%). The 2021 build: 1,560,000 − 780,000 = 780,000 — corroborating, but fragile (improvement estimate is half the price; a 20% error swings the land ±$156,000): down-weight it. Allocation check: 80% × ~$975,000 typical improved value ≈ $780,000 — consistent. Reconcile: ≈ $790,000–$795,000 per lot. Extraction led, allocation corroborated, and the weighting followed each method's error math.
Common exam pitfalls
Extracting from new construction.
When improvements dominate the price, the residual inherits every estimating error — extract from low-contribution sales.
Importing land ratios across market types.
Ratios are market-, class-, and cycle-specific — a suburban 25% has nothing to say about beachfront 80%.
Leading with allocation when sales exist.
Sales comparison first, always; extraction next; allocation corroborates.
Subtract the building to find the dirt; ratio the town when even that fails — and trust the method whose errors are smallest.
Recap
- Extraction: improved price − depreciated improvement cost
- Best where improvements contribute least
- Allocation: typical land ratio × total value
- Ratios are local, class-specific, cycle-sensitive
- Hierarchy: sales comparison → extraction → allocation
- Stack and reconcile for credibility
Prove it: 10 questions on this topic
Every lesson ends with a ten-question check in the free course — your progress syncs between the web and the EstatePass app.
Studying for the appraiser licensing exam? Track every lesson free — progress syncs with the app.
Start freeMore in Site Valuation
Study smarter in the free dashboard
- Every lesson tracked, synced with the iOS app
- Ten-question checks after each lesson
- Lesson videos, flashcards and mock exams
No credit card required.