The Mass-Appraisal Feedback Loop

The structural finding in one sentence

Under Washington statute, county assessors must value property at "true and fair value in money" (RCW 84.40.030). In practice this is delivered via mass appraisal: a model that assigns land values to thousands of parcels at once using a small set of actual sales plus neighborhood land-rate tables. Those land-rate tables capture local supply-demand conditions — but the supply side of that equation is heavily determined by the same city/county government's zoning, UGA boundary, lot-size minimums, parking minimums, and permit-pipeline policies. The assessor uses the price effects of its own employer's land-use restrictions as the inputs to the valuation model that determines that employer's tax revenue.

The mechanic: how mass appraisal works

Whatcom County (and every WA county) assesses property in three layers each year:

• Land value: derived from neighborhood land-rate tables (so much per square foot of land in a given location, adjusted for waterfront / view / corner / shape / topography characteristics). Tables are calibrated against the small number of vacant-land sales that occurred plus "extracted residuals" from improved-property sales (sale price minus depreciated improvement cost = imputed land value).
• Improvement value: derived from cost-replacement-new minus depreciation (Marshall & Swift cost manuals), occasionally cross-checked against direct sales of comparable improvements.
• Total assessed value: land + improvements. This is what the tax rate multiplies against.

For most parcels in any given year, no direct sale has occurred. The land-rate table is the source of the assessed land value. The table changes annually based on whatever sales DID occur — which is a small fraction of total parcels.

The empirical reality inside Bellingham city limits

The Whatcom County Assessor's roll lists approximately 2,700–3,200 parcels with zero improvement (vacant land) inside Bellingham city limits in any given year. Most are not actively for sale — long-term holdings, easement-restricted, institutional (church-owned, school-owned, parks-owned, utility-easement), or unbuildable due to steep slope / wetlands. The relevant question is not the count of vacant parcels but the count of vacant-land transactions the assessor can actually use to calibrate a model.

Per Northwest Multiple Listing Service® (NWMLS InfoSparks), closed vacant-land sales in the Bellingham NWMLS area by year:

Two patterns are visible in the data:

1. Total annual vacant-land sales are small. Even in peak years (2006, 2021) the Bellingham NWMLS area saw under 200 transactions. The assessor's mass-appraisal model values 17,609 single-family-residential parcels inside city limits against this small set of transactions, supplemented by extracted-residual values from improved-property sales.
2. The 2020–2025 supply-side trajectory ran the opposite direction from prices. Vacant-land sales fell from 152 → 80 over the 5-year period (a 47% decline) while land values approximately doubled. In a normal market, higher prices attract higher transaction volume as owners realize gains; in Bellingham, sales volume halved during the boom.

The arithmetic of mass appraisal means that when a shrinking number of vacant-land sales price high (because they're rare and demand is constrained), every other parcel's imputed land value rises with them — regardless of whether that other parcel has been sold or even put on the market. Existing homeowners pay tax on imputed appreciation, not realized gain.

Data caveat: NWMLS InfoSparks reports the Bellingham geographic area, which approximates but does not exactly match City of Bellingham municipal boundaries. Closed sales include all vacant-land categories reported through NWMLS; private/off-market transfers and FSBO sales not transacted through the MLS are not captured. Real Record has filed a PRR to the County Assessor for the exact vacant-residential-land transaction count inside city limits used in mass-appraisal calibration.

The structural incentive

The principal-agent question:

• City of Bellingham (legislative authority) sets zoning, lot-size minimums, parking minimums, permit pipeline, and through-zone density rules.
• Whatcom County (separate legislative authority) sets the UGA boundary that constrains where city-style development can occur.
• Both governments collect the property tax that results from the assessed values produced by mass-appraisal methodology.
• The mass-appraisal model captures the price effects of those land-use restrictions as inputs.
• Restrictive land-use rules increase modeled land values (because supply is constrained) → assessed values rise → tax revenue rises.

None of this is hidden. The mass-appraisal methodology is documented public practice (see Washington State Department of Revenue's Property Tax Division guidance for county assessors). What it amounts to as a system is: the same government that determines whether more land becomes available for housing also benefits financially from keeping land scarce.

The Bellingham demonstration: supply response ran the wrong way during the boom

Real Record's sister investigation — The Bellingham Small-Lot Tax Premium — quantifies what this looks like in practice. Over the 2020–2025 period (the COVID-era housing-price boom), Bellingham land values inside city limits roughly doubled for standard residential lot sizes. In a normal supply-elastic market the price signal would have attracted dramatic new supply. Instead:

• 5,000–7,500 sq ft (standard residential) new-lot creation fell from 29.2 to 12.8 lots per year — a 56% drop during the boom.
• 7,500–10,000 sq ft new lots fell from 16.2 to 7.6 per year — also a 53% drop.
• Townhouse / zero-lot-line (under 2,500 sq ft) new-lot creation rose substantially, but from a small base — the only category where current zoning permits new density.
• Total new-lot creation across all sizes fell 19% (108/yr pre-boom → 87/yr during boom).

The model that captures these price increases continues to assign higher and higher land values to every parcel that didn't sell — not because the market validated those prices through transactions on each parcel, but because the model inherited the inputs the regulated environment generated.

What this would look like if it weren't a feedback loop

In a system where the assessor's inputs were genuinely independent from the assessor's employer:

• Land-rate tables would be normalized against unrestricted supply (e.g., calibrated against an idealized non-zoned land market) — or at minimum, the model would treat supply-constrained jurisdictions differently from supply-elastic ones when applying scarcity premia.
• Periodic re-baselining would reset the model when policy changes alter supply (e.g., upzoning would reset land-rate tables downward to reflect the new supply environment, not just continue increasing).
• An assessment-cap mechanism (similar to California Prop 13, or Washington's I-747 1% growth cap but applied to assessed values not levies) would limit the year-over-year increase in imputed value, regardless of model output.

None of these mechanisms are currently in WA state law. Assessed values can rise without limit; only the LEVY rate is capped at 1% growth (under I-747). When values double in 5 years, levy rates adjust downward to keep total-collection growth at 1% — but the tax burden shifts across parcels. Owners of parcels whose modeled values rose fastest pay a disproportionately larger share.

Editorial frame

This is not a claim that any individual assessor or assessment is wrong. The mass-appraisal methodology is professionally accepted, the math is defensible per RCW, and assessors are following the statute. The question is whether the statutory framework itself creates a structural incentive against supply expansion: when more restrictive land-use policy yields more property-tax revenue without legislative-body action (no levy rate increase, no public vote), the political economy tilts in favor of keeping policy restrictive.

Real Record reports the methodology and the structural incentive. Readers may conclude:

• The system is fine and the modeled values reflect genuine market scarcity, OR
• The system contains a perverse incentive that quietly funds government from the side-effects of its own restrictive policies, OR
• The methodology is sound but the framework needs reform (assessment caps, periodic re-baselining, decoupled inputs).

The information is there; you decide.

Research status

This is an active investigation. Real Record is gathering:

1. The actual count of vacant-residential-land transactions inside Bellingham city limits per year 2016–2025 (PRR pending).
2. The Whatcom County Assessor's neighborhood land-rate tables and the methodology documents that produce them (PRR pending).
3. Internal Assessor correspondence about land-rate-table adjustments during the 2020–2023 boom period (PRR pending).
4. Per-parcel year-over-year assessed-value changes for parcels that did NOT sell, to quantify how much of the average homeowner's tax-bill increase is imputed-model output vs. market-validated.
5. Equivalent comparisons from a supply-elastic comparison jurisdiction (a Texas or Houston-area county with similar growth and far fewer land-use restrictions).

The investigation will be updated as findings arrive.

1. Whatcom County Assessor: All neighborhood land-rate tables, methodology documents, and calibration notes used for the 2020, 2021, 2022, 2023, 2024, and 2025 rolls for parcels inside Bellingham city limits.
2. Whatcom County Assessor: All actual vacant-land transactions inside Bellingham city limits 2016–2025, with sale price, buyer/seller identity (to the extent recorded), and the assessor's annotation of whether each sale was used to calibrate land-rate tables.
3. Whatcom County Assessor: Internal correspondence (email, memos, meeting minutes) discussing land-rate-table adjustments during the 2020–2023 housing-price boom.
4. Washington State Department of Revenue, Property Tax Division: Methodology guidance issued to county assessors 2018–2025 regarding mass appraisal in supply-constrained jurisdictions.
5. City of Bellingham: All staff reports and Planning Commission / City Council minutes 2018–2025 discussing the projected fiscal impact of zoning changes (upzones, downzones, minimum-lot-size changes) on property-tax revenue.

Source for mass-appraisal methodology: Washington State Department of Revenue, Property Tax Division — assessor guidance documents; RCW 84.40.030 (true and fair value in money standard); Marshall & Swift cost manuals as referenced by WA county assessors for improvement-value cost approach.

Source for the supply-response data: Whatcom County Assessor tax_parcels + tax_assessments tables, 2016–2025 roll years, filtered to inside-Bellingham-city-limits parcels via tax_area_description LIKE 'BELLINGHAM%', single-family-residential and zero-lot-line use codes (1111–1115, 1133). Methodology details in The Bellingham Small-Lot Tax Premium.

Source for vacant-parcel counts: Whatcom County Assessor 2016–2025 rolls, filtered to inside-city Bellingham parcels with market_improvement_val = 0 and market_land_val > 0. Yields ~2,700–3,200 vacant parcels per year, dominated by easement-restricted, institutional, or long-held parcels not actively offered for sale.

Caveats: (1) Mass appraisal is a professionally accepted methodology and is not, by itself, evidence of wrongdoing. (2) Individual assessors are constrained by RCW to use this approach and are not the policy actors who set land-use restrictions. (3) The structural-incentive argument is about the framework, not the practitioners. (4) Counterfactual claims ("what would land values be without the restrictions") are not testable from in-jurisdiction data alone; cross-jurisdictional comparisons are pending.