Income tax calculator methodology

What we calculate

Our default scenario is a full-year tax resident earning ordinary employment income from one employer. The engine subtracts a standard personal allowance, applies marginal bands progressively, adds common employee-side mandatory contributions and any selected regional estimate, then reports net pay.

Employer payroll contributions are calculated separately. They increase employer cost but do not come out of contractual gross salary.

Detailed and screening models

Global breadth and local precision are not the same thing. We use two visible model labels and treat them differently in publication:

Detailed model

Country-specific tax assumptions with linked authority material. Some systems use approximations rather than their complete statutory formulas. Detailed does not mean return-ready: credits, household rules and unusual income can still be excluded. Only these 18 models receive indexable country pages, ranking positions and API records.

Illustrative scenario

A directional screening model available only by opting in inside the interactive calculator. It uses a country headline rate and contribution estimate, then creates broad illustrative thresholds from the stored salary baseline. Those thresholds are not statutory brackets and must not be read as local tax law. Illustrative results are excluded from published rankings and country landing pages.

There are currently 18 detailed models and 96 screening models. The screening set helps users decide where deeper research may be useful; it is not presented as sourced tax guidance.

Progressive tax bands

Each rate applies only to the slice of taxable income inside its band. Moving into a higher band does not cause all prior income to be taxed at the higher rate. The calculator exposes the band-by-band arithmetic inside each result card.

Formula-based systems, credits and tapered allowances may require a model-specific adjustment. Country pages disclose those approximations.

Effective rate, marginal rate and tax wedge

Effective deductions are income tax, regional or local tax and employee contributions divided by gross income. Marginal rate is measured, not looked up: the engine recalculates the scenario with 100 more units of local gross pay and divides the extra deductions by 100. That way allowance tapers, contribution ceilings, credits, surtaxes and rebate thresholds all show up, rather than only the statutory band rate.

Tax wedge follows the OECD Taxing Wages definition: employee deductions plus employer contributions, divided by total employer cost (gross salary plus employer contributions). Employer contributions are modelled as one average rate on gross pay, without ceilings or sector rates, so employer figures are orders of magnitude rather than payroll quotes.

Pay-period breakdowns divide annual results by 12 months, 26 fortnights, 52 weeks, 260 working days or 2,080 paid hours. They describe an annual average, not the timing of an individual payslip.

Currency conversion

For an international comparison, the engine converts gross income from the selected input currency into each country’s local currency. It applies local thresholds there, and only then converts results back to the comparison currency.

Our planning exchange-rate table is dated 4 September 2026. It is a stored set of assumptions, not a live quote or verified historical market series. Edit each rate in the calculator and include it in a shared scenario. In offer mode, salaries stay in local currency and FX affects their comparative presentation. In equal-salary mode, FX also changes the local gross salary to which tax is applied.

Salary examples and living costs

Example salary baselines are internal planning inputs. No statistical source or observation period was recorded for them, so they are labelled “example salary”, not average or typical earnings. Their provenance was audited on 30 September 2026; that date does not make them observed wage data.

The previous stored national cost indices likewise have no documented statistical source or observation year. We no longer publish purchasing-power rankings from those values. References about PPP explain the concept; they do not validate those stored numbers.

In “Different job offers”, enter your own monthly rent, healthcare, childcare, transport and other spending as one local-currency budget. We subtract that budget from annual net divided by 12 and convert the remainder to your comparison currency. A zero budget means costs have not been supplied. This is a personal scenario, not an official PPP measure.

What we leave out

• Tax-residence determination and treaty tie-breakers
• Non-salary income, capital gains and wealth taxes
• Most credits, dependants and itemised deductions
• Municipal tax unless explicitly stated
• Benefits in kind and equity-compensation timing
• Immigration, permanent-establishment and payroll-registration analysis

Sources, updates and corrections

Detailed country pages link to the authority material used and show a review month. Model dates describe the stored dataset. They do not certify completeness or indicate that every linked authority has been rechecked on a product release date.

Updates are made in the country data file, exercised against regression tests, and then reflected in country pages, comparisons and rankings from the same calculation engine. This avoids maintaining conflicting copies of the same rate.

If you find a discrepancy, include the country, tax year, input scenario, expected result and primary source in your correction note. Reproducible reports are prioritised. See the change history for the scope of each update.