How we calculate your numbers

Every figure in a Reloc8te report is either verified — computed by one of our calculators from a named government or commercial data source — or clearly marked as an AI estimate. This page explains which is which, where the data comes from, how the headline savings number is built, and what it deliberately leaves out.

Applies to

Reports generated with the verified savings model (“How we got this number” panel present)

Last updated

September 2026

Tax tables

Federal and state income-tax tables — see the tax year shown in your report (the tables in force when it was generated)

Questions

support@reloc8te.com

What’s verified vs approximate

A report mixes three kinds of content. They are labeled differently on purpose, and the label tells you how much weight to put on the figure.

Data sources & refresh cadence

External data is fetched per city, cached, and refreshed on the schedule below. Where a city-level figure is not published we fall back to the county or state figure and say so in the report (for example “statewide index” on the cost-of-living tile). We do not fill gaps with guesses.

All third-party data may be delayed, revised, or contain errors, and is not independently verified by Reloc8te. Vintages are shown next to the figure where they matter (for example the BEA RPP year).

How the savings number is computed

The headline “Approximate Annual Savings” (or “Added Cost”) is the sum of the lines shown in the report’s “How we got this number” panel — nothing else contributes. Each line is computed on our servers by a calculator, not written by the AI model, and each carries its source and assumptions. Every destination in a report gets its own total, built the same way.

Income tax difference

• Computed by our tax engine from your household income, filing status and (where provided) itemized income sources, using the federal and state tables for the tax year shown in your report.
• Federal tax, FICA and self-employment tax are included so the federal side is consistent between the two locations.
• State tax for graduated-rate states uses a blended effective rate: the top marginal rate that applies to your income × 0.78. Flat-tax and no-income-tax states use their actual rate.
• Not included: local or city income taxes (for example New York City or Ohio municipalities), credits specific to your return, and state-specific treatment of retirement or military income beyond what the engine models.

Property tax difference

• Destination: Zillow ZHVI metro-typical home value × the county’s effective property-tax rate (Census ACS median real-estate taxes paid ÷ median home value). Origin: your stated home value (or the ZHVI typical value) × your county’s rate. Where no county figure exists, the state-average rate table is used and the line says so.
• County medians reflect what current owners pay; a new purchase at market value can run higher, and school-district and municipal rates vary within a county. Each line’s assumptions name the exact rates used.
• Homestead exemptions are not applied in the line (they vary by county and require application); they are called out in the narrative where relevant.
• If only a statewide home value exists for a destination, the property-tax line is omitted and listed under “Not included in this total” rather than computed from a guess.

Veteran property-tax exemption

• Computed from each state’s rules (full, tiered, proportional, credit or percentage-of-tax) applied to your VA disability rating and the destination home value.
• Exemptions must be applied for, usually with the county assessor or appraisal district, and some are income-tested; the line’s assumptions say so.
• Surviving-spouse carry-over is modeled where state law provides it; remarriage and cross-state moves generally end it.

What is NOT in the total, and why

• Cost of living — the percentage stays the measured comparison (see below); the index itself is never converted to dollars. If you enter your monthly figures, an approximate living-cost change is computed separately from them and shown beneath the total, never inside it (see approximate living-cost change).
• Sales tax — the combined state + local rate is compared in the Taxes chart, but how much your household spends is not measured, so there is no honest dollar figure.
• Insurance, utilities, groceries, fuel, childcare — not in the verified total. When you enter your own monthly figures they are approximated separately, scaled by public price data for each place (see below). HOA dues and moving costs stay unmeasured for either location.
• Rental cash flow from a home you keep is computed but shown separately: it does not change with where you move, so adding it would inflate every destination equally.
• One-time equity (your current home value minus the destination typical value) is shown separately and labeled one-time, never annual.

Whenever a line could not be computed, the report lists it under “Not included in this total” with the reason in plain language — that list is how you know what the headline leaves out.

Cost of living — why a percentage

We compare overall price levels using the BEA’s Regional Price Parities, where the U.S. average is 100. A destination at 114.2 against an origin at 98.6 is about 15.8% more expensive overall. That percentage is a measured, published comparison. Turning it into “$X per year” would require knowing how much your household spends on price-sensitive goods — a figure we do not ask for — so any dollar conversion would be an assumption dressed as a measurement. The report therefore shows cost of living as a percentage tile, keeps it out of the dollar total, and notes when the index is statewide rather than metro-level.

The one place dollars and price levels meet is the approximate living-cost change, and only because the spending base there is yours, not an assumption of ours. That figure is shown separately from the verified total and is never added into it.

Approximate living-cost change — your figures × public price data

The verified total covers taxes. The largest everyday differences between two places — groceries, fuel, utilities, insurance, childcare — depend on how much your household spends, so we ask for that instead of assuming it. The result is labeled Approx. living-cost change, carries an estimate chip rather than a Verified one, and sits beneath the total, never inside it.

What you enter

• Five categories — groceries, fuel, utilities, homeowners insurance and childcare — as monthly dollar amounts. All are optional; skip any you do not know.
• For each category you can accept a preset instead of typing a figure. Presets come from national federal tables and are labeled typical everywhere they appear; a preset never pretends to be your own number.
• The block can be switched on or off in Settings → Housing. Off, or with nothing entered, the report shows no living-cost figure at all and says so.

The formula

• For each category: destination annual = your monthly × 12 × destination index ÷ origin index. Only the ratio between the two public figures is used; the index itself is never turned into dollars.
• Change = origin annual − destination annual. A positive amount means that category costs less at the destination; a negative amount means it costs more.
• A ratio outside 1⁄3–3 is treated as a data problem rather than a real difference: the category is omitted and listed under “Not estimated” with that reason.
• The category amounts are summed into one living-cost change per destination. Each line shows its two index values, their grain and vintage, and the arithmetic in words.

Where the price data comes from

Presets and confidence

A preset you accepted is labeled typicalon its line and lowers the confidence of the figure, because a national average says little about your household. Confidence for the living-cost change is earned the same way as elsewhere — by what was entered and how finely the data resolved — and is never self-reported:

High: at least three categories entered by you and priced at ZIP, metro or county grain in both places. Medium: at least one figure entered by you (a statewide or regional average on either side holds that category at medium). Low: typical presets only. The combined net takes the lower of the two confidences.

The combined figure

When a living-cost change exists, the report also shows Combined approximate net: the verified tax savings plus the approximate living-cost change, at the lowerof the two confidences. It is a convenience line so you can see both effects together. It never replaces the verified headline, never enters the “How we got this number” breakdown, and moving the What-If sliders changes only the verified part — your stated spending is not rescaled by income or home price.

Caveats

• Your own bills beat any estimate. If you know what you pay today and can get a quote for the destination, use those; this figure exists to show direction and rough scale where you cannot.
• Homeowners-insurance premiums are the Treasury FIO’s 2018–2022 data (2022 is the latest year in the file); premiums have moved since, and the line says which year it used.
• ZIP-level premiums are used only where the FIO dataset has enough reporting insurers and policies for that ZIP; otherwise the state figure stands in and the line says so.
• Presets are national. They are a starting point for the arithmetic, not a description of your household, which is why they lower confidence.
• A statewide or regional index on either side cannot see differences between neighborhoods, and every source lags by a year or more.

Confidence levels

Confidence is earned by the calculators, not self-reported by the model:

High

Both the income-tax line and the property-tax line were computed from verified inputs.

Medium

One of the two was computed; the other is listed under “Not included” with its reason.

Low

Neither could be computed, or an automated accuracy check failed. When a check fails, the report lists the findings under the confidence pill (“Why confidence is low”).

What “AI estimate” means

The narrative parts of a report — section summaries, neighborhood suggestions, lifestyle-match scores, alternative cities — are written by an AI model from the verified data plus its general knowledge. Where the model supplies a number itself, it is stamped AI estimate in amber and carries no source. These figures are unverified and should be treated as context to check, not as data. Alternative cities never show a savings figure, because none was computed for them.

Known limitations

• Tax tables follow the tax year shown in your report; a later year’s brackets, standard deduction and credits may differ.
• State income tax uses a blended effective rate for graduated states rather than your full state return; local income taxes are excluded.
• Property tax uses metro-typical home values and county-median effective rates (state-average where no county figure exists); your actual bill depends on the specific parcel, taxing districts, purchase price and exemptions.
• Cost-of-living indices are metro- or state-level and lag by a year or more; they cannot see differences between neighborhoods.
• Crime, school and hazard data have reporting gaps and are relative indicators, not predictions.
• Home values and rents are market estimates that move monthly.
• Veteran exemptions change by legislative session and usually require a timely application; confirm with the county before relying on one.
• Reports generated before the verified savings model was introduced show a model-authored breakdown (tagged as such) or none at all.