The 19% Flat Rate: When It Beats the Brackets

“Foreigners pay 19%” is not what this provision does.

It lets a qualifying foreign employee elect 19% on Korean employment income instead of the ordinary 6–45% scale — 20.9% once local income tax is included — in exchange for giving up, as a rule, the earned income deduction, personal deductions, special deductions and most credits.

Which means the decision is not about your salary. It is about what those deductions were worth to you.

Compare total tax, not 19% against a bracket number.

Whether you are a Korean tax resident at all, and what that puts in scope, is the prior question — answered in income tax for foreigners: resident or not. This guide assumes that is settled and deals only with the election.

19% of income, not of a tax base

The two methods do not differ only in the rate. They differ in what the rate is applied to.

Progressive   gross pay
              → earned income deduction
              → other deductions
              → tax base
              → 6–45%
              → tax credits

Flat          employment income
              → 19%

Reading the flat rate as “tax base × 19%” is the expensive mistake, because under the progressive method the base is a good deal smaller than the gross before any rate touches it.

What the election gives up

Electing the flat rate sets aside, in principle, the exemptions, deductions, reductions and tax credits under the Income Tax Act and the Restriction of Special Taxation Act. That includes the most basic one of all — the earned income deduction — along with:

• basic and dependant personal deductions
• special deductions such as insurance premiums and housing funds
• the credit card usage deduction
• the child tax credit
• medical, education and donation tax credits
• the monthly rent tax credit

Side by side

Comparing 19% against 24%, 35% or 38% directly is what produces the wrong answer. Under the progressive method those rates apply band by band, not to the whole salary, and only after the earned income deduction has come off.

The modelled crossover

Take a single foreign employee with no dependants, the basic deduction, the earned income deduction and the earned income tax credit, and no material card, rent, insurance or donation deductions. Local income tax counted on both sides.

What it does establish firmly is a negative: the frequently repeated claim that the flat rate starts winning somewhere between ₩70m and ₩100m is wrong, and following it costs a mid-career employee millions of won.

Every deduction pushes the crossover higher

Two employees on ₩150,000,000 can reach opposite conclusions.

A is single, no dependants, negligible rent, education, medical or donation deductions. The flat election starts paying off quickly.

B has dependants, qualifying housing-related deductions, and meaningful medical, education, donation and card deductions. Every one of those reduces the progressive tax base and none of them touches the flat calculation — so B’s break-even sits far above ₩135m, possibly above any salary B will realistically earn.

Which is why the decision rule is not a salary threshold at all:

Is the tax saved by the flat rate
  >
the value of the exemptions, deductions and credits you give up?

Answer that and the salary figure becomes an output rather than an input.

At the top the gap opens fast

Past roughly ₩150,000,000 the progressive tax base starts occupying the 35% band and then 38% and above, while the flat calculation stays at 19%.

Modelled, ₩150m favours the flat election by around ₩2.5m; ₩200m by around ₩12m. So for a senior expatriate executive or a highly paid professional with few Korean deductions, running the comparison is worth real money.

The reverse warning matters just as much: being in the 35% band is not a reason on its own. Middle-income employees frequently sit in a high marginal band while their average rate stays well below 19%.

Two applications, two moments

The provision has a payroll layer and an annual layer, and conflating them causes unnecessary alarm.

Missing the first does not forfeit the provision. It changes your cash flow through the year — more withheld each month, settled later — but the final tax position is decided at step two.

Waiving it takes effect next year

If you have flat-rate withholding running and your circumstances change, the waiver application takes effect from the tax period after the one in which it is filed.

So telling payroll in June to switch you back from July is not how it works. Whether the year as a whole ends up on the flat rate or the progressive scale is settled at year-end settlement or on the return — which is the argument for modelling both in January rather than discovering the answer in December.

What payroll should actually run

Not one calculation in December. Two, in January, refreshed when anything moves.

The band worth watching closely is ₩120,000,000 to ₩150,000,000, where the two methods run close enough that a bonus decides it.

The National Tax Service’s own year-end settlement service compares flat against progressive directly. Use it against real payroll figures rather than any threshold quoted in an article — including the one in this table.

The decision, restated

The flat rate’s value is not simplicity. It is capping the part of a large salary that would otherwise be taxed at 35% and above.

Its cost is the exemptions, deductions and credits you stop being able to use.

Compare those two, not the salary against a threshold. On a single filer with nothing to deduct, the model says the line is near ₩135,000,000 — and your own line is somewhere else, higher, by exactly as much as your deductions are worth.

What the settlement itself involves, and what a foreign employee can and cannot deduct under the ordinary method, is in income tax for foreigners: resident or not.