Property Tax Caps

What the evidence says

Assessment caps work through a specific mechanism: they hold a property's taxable value below its market value for as long as the current owner stays put, and that gap resets to market value only at sale. California's Proposition 13 sets this up literally — a 1% rate cap plus a 2%-per-year assessment growth cap — and the measured consequence is a lock-in effect on mobility. Owner tenure length in California rose by 1.04 years and renter tenure by 0.79 years relative to comparison states between 1970 and 2000. The renter effect runs through a second-order channel: fewer owner-occupied units turn over, so households spend more time renting before they buy.

The size of the lock-in effect is not uniform — it scales with the dollar value of the subsidy an owner is sitting on. In Bakersfield, where the average Prop 13 subsidy is $110/year, tenure length rose by only 0.11 years. In Los Angeles/Orange County, a $1400/year subsidy produced a 2.3-year increase; in San Francisco/San Jose, a $3500/year subsidy produced a 3.0-year increase. Coastal cities generally, with subsidies in the thousands of dollars, show 2–3 year tenure increases. This dose-response pattern comes from one research program (Wasi), published first as an NBER working paper and later in Brookings-Wharton Papers on Urban Affairs with the same underlying data and figures.

The same research also finds the lock-in effect is not evenly distributed by race or origin. African-American households show a larger response than white households (a difference-in-difference-in-differences of 1.40 years for owners and 1.96 years for renters in the NBER version; 1.3–1.8 additional years in the published version). Out-of-state migrant households show a larger response than native-born households (3.37 years for owners, 0.77 years for renters) — this migrant comparison appears in only one source in this set and is not cross-checked elsewhere.

At the aggregate level, California's effective property tax rate on single-family homes with FHA mortgages fell from .0221 to .0055 (a 75% decline) between 1977 and 1987, compared with a 31% decline (.0167 to .0115) in the US overall over the same period. A separate, lower-confidence modeling source reports that rate limits on their own decrease revenue only in their year of enactment, with no measured effect on growth in later years, and that assessment limits and rate limits each fail to constrain long-run growth unless combined — consistent with Prop 13 pairing a rate cap with an assessment cap rather than using either alone. That same source reports property tax limits decrease state and local revenue during recessions, with larger declines in states that lean more on income-elastic revenue sources, though this claim carries medium/low confidence and comes from an unauthored source.

At the school-district level, Indiana's caps (1% residential, 2% agricultural/rental, 3% business) produced 2016 funding losses ranging from under $100 to over $17,000,000 per district across 292 districts, totaling $238,359,015 (mean $824,772, median $97,864). Loss size increases with enrollment (r=0.696, p<.001) and per-pupil levy (r=0.190, p=.001), and decreases with per-pupil assessed valuation (r=-0.143, p=.015); enrollment and levy together account for 50.9% of the variance in losses. Losses also track district demographics: higher shares of students of color (r=0.576, p<.001), English learners (r=0.322, p<.001), and students receiving free or reduced-price meals (r=0.277, p<.001) each associate with larger losses, while special education share shows no measured relationship (r=-0.008, p=.751). Districts that received state emergency debt-service waivers because of these losses were disproportionately urban (58%) and suburban (44%), rarely rural (10%). Separately, one source reports property tax caps increase local adoption of growth-control policies, on the reasoning that capped assessments make new residents comparatively less attractive to existing owner-voters — though this claim states direction only, with no quantified magnitude.

Where the evidence splits

No contradictions in the current sources.

Thresholds and dose-response

The lock-in effect on owner tenure scales continuously with subsidy size rather than switching on at one cutoff: $110/year subsidy → 0.11 years added tenure (Bakersfield); $1400/year subsidy → 2.3 years added tenure (LA/Orange County); $3500/year subsidy → 3.0 years added tenure (SF/San Jose); coastal cities generally, thousands-of-dollars subsidies → 2–3 years added tenure.

Rate limits show a time-based threshold rather than a magnitude one: they decrease property tax revenue in the year of enactment, but that effect does not carry into subsequent years, and neither rate limits nor assessment limits alone reliably constrain growth once that first year passes.

Open empirical questions

• Whether Florida's more lenient 3%-or-inflation assessment cap produces a measurable mobility or revenue effect comparable to California's 2% cap — these sources describe Florida's design but report no outcome data for it.
• Whether school-district revenue losses from property tax caps persist, grow, or shrink in years after 2016 — only a single year is measured.
• What magnitude of growth-control policy adoption results from property tax caps — the source establishes direction only, with low confidence in any effect size.
• Whether the renter tenure-length increase operates through channels other than delayed homeownership — only the mechanism is stated, not separately decomposed from the reported 0.79-year figure.