What tax bracket am I in, and what is my real rate?

A tax bracket is a fact about one line of Form 1040. It is the rate that applies to the last slice of taxable income on line 16, and at $210,000 of wages filing head of household it is 24.0%. Every bracket calculator on the internet will tell you that, and it is correct.

The rate on your next $1,000 is a fact about the whole return. Earn $1,000 more and line 16 rises by $240, exactly as the bracket promises. But the child tax credit on line 19 falls by $50, and the additional Medicare tax on line 23 rises by $9. Total tax on line 24 goes up $299. You keep $701 of the $1,000. That is a real rate of 29.9%, which is 5.9 points above the bracket you would have quoted.

Both numbers are computed the same way here, and neither is looked up. The calculator runs a complete Form 1040 at your income, runs it again $1,000 higher, and subtracts. The bracket is the change in line 16 over the increment. The real rate is the change in line 24, plus anything the payments on line 33 gave up, over the same increment. A table lookup cannot produce the second number, because the second number depends on credits the table does not know you have.

Three rules can charge your next dollar without appearing in a bracket table, and this page prices each one separately rather than folding them into a single rate.

The child tax credit under § 24 begins to disappear once income passes a threshold, at $50 of credit per $1,000 of income. At the inputs above that costs $50 of the $1,000, which is 5.0% on top of the bracket, and it applies until the credit is exhausted. You can see its size by taking the children away: the identical household with no children faces 24.9% on its next $1,000 rather than 29.9%, so the credit phase-out is 5.0 points of the 5.9 points gap.

The additional Medicare tax under § 3101(b)(2) adds 0.9% on wages above the threshold for your status, which is $9 of the $1,000 here. It is the remaining 0.9 points in the childless case. It is charged on wages, so it fires on a raise and not on interest, dividends, or a retirement distribution. That is why the calculator also prices the same slice as income that is not wages: 29.0% instead of 29.9%.

The earned income credit under § 32 is the third, and it does not touch a tax line at all. It sits on line 27a as a payment, and it shrinks as earnings rise. A household in that range loses real money on a raise while its total tax stays exactly where it was. The section below is that case.

The gap is not a constant, and it is not a high-earner phenomenon. It opens where a credit starts to disappear and closes once the credit is gone. Same household throughout: head of household, 2 children under 17, standard deduction, wage income only, every row two complete returns rather than a formula.

At $80,000 the bracket is honest: 12.0% and 12.0% agree, because nothing is phasing out. The gap is widest at $210,000, where a 24.0% bracket hides a 29.9% rate. By $400,000 the child tax credit has run out entirely, so there is nothing left to withdraw and the gap falls back to the 0.9% of additional Medicare tax. The reader who most needs this page is not the one at the top of the table. It is the one in the middle of it, being told they are in the 24.0% bracket.

Take a head of household filer earning $45,000 with 2 children under 17. The standard deduction of $23,625 leaves $21,375 of taxable income and $2,225 of tax, which the child tax credit erases completely. Total tax for the year: $0. The earned income credit adds $2,587 on the payments side.

Now add $1,000 of wages. Line 16 rises $120, the child tax credit absorbs all of it, and line 24 finishes where it started at $0. On the tax lines alone the raise was free. But the earned income credit falls from $2,587 to $2,377, so total payments on line 33 drop $210. The household is $210 worse off on $1,000 of extra pay. That is 21.0% against a 12.0% bracket, the largest gap anywhere on this page, and it lands on the lowest earner in it.

This is worth being precise about, because it is where calculators go wrong in a way that matters. The tax engine behind this page exposes an incremental-rate projector, and asked for the federal rate on this exact case it returns 0.0%. That answer is correct for the question it was asked, which is how much more tax is owed. It is the wrong question. This page asks how much less money the household ends up with, so it subtracts the change in line 33 as well, and reports 21.0%.

People use three rates interchangeably and they are not the same. Your bracket is 24.0%. The next $1,000 costs 29.9%. And the year as a whole cost 15.25%: $32,028 of federal income tax on $210,000 of income.

The effective rate is the smallest because most of your income was never taxed at your bracket. The standard deduction sheltered the first $23,625. What remained was cut into slices and charged at rising rates, so only the last slice met 24.0%. Credits then came off the total. That is why $35,838 of tax before credits became $32,028 after them.

Which one you want depends on the decision. Deciding whether to take on extra work, sell something, or negotiate a bonus is a marginal question, and 29.9% is the number. Deciding whether you can afford something out of this year's income is an effective-rate question, and 15.25% is the number. Quoting your bracket, 24.0%, answers neither one exactly.

The refundable child tax credit. When credits exceed the tax they offset, part of the excess comes back as the additional child tax credit on Schedule 8812 under § 24(d). This engine computes the child tax credit but does not carry the refundable part to Form 1040 line 28. In the head of household case above, $4,400 of credit meets $2,225 of tax, so $2,175 has nothing left to offset and the refundable calculation is live. That amount moves with income, so the real rate for a household in that position is not exactly 21.0%. The calculator raises a warning naming the unused amount rather than printing a number it cannot stand behind, and the children input is capped at 2 because at three the refundable formula changes to one this engine does not compute.

The 3.8% net investment income tax under § 1411 and the qualified business income deduction under § 199A. Both can move a marginal rate sharply, and both need income this page does not accept: NIIT applies to investment income above a threshold, and § 199A applies to pass-through business income with its own phase-in on wages and property. This page models wage income only, so neither can fire here. If your next dollar is a dividend, a capital gain, or a distribution from a business you own, this is the wrong page for it.

Everything outside federal income tax. State and local income tax is on top of every number here and in some states adds more than a full bracket. Social Security and Medicare withholding come out of the same check under separate rules, and only the 0.9% additional Medicare surtax is included. Self-employment tax, itemized deductions, retirement contributions that reduce adjusted gross income, and the many other credits with their own phase-outs are all absent. The parameter set is tax year 2025 only, not a history of past brackets, and none of this is tax advice.

This calculator is built to be used without a browser. Every input is a query parameter on this page, and the same parameters on the JSON twin return the complete answer as a document.

GET /tools/marginal-rate.json?income=210000&status=head-of-household&dependents=2

The response carries inputs after parsing and clamping, result with the statutory bracketRate, the true marginalRate, the gapPoints between them, the effectiveRate, and raise, the increment split into the rate schedule, credit phase-outs, other taxes, and refundable credits withdrawn, plus otherIncome for the same slice priced as income that is not wages, engineOrdinaryRate from the engine's own incremental projector, form1040 as a before-and-after line walk, sensitivity across an income ladder and all five filing statuses, and computability, the engine's verdict on both returns, drivers ranked by effect with a plain sentence each, assumptions that say for every field whether you supplied it and name the source when the default came from one, sources with a URL and an as-of date for each source the tool cites, which is an empty list on the calculators whose every default is an example input rather than a published figure, warnings, a disclaimer, and in tool the canonicalUrl and jsonUrl that carry only your non-default parameters. The canonical URL is the answer's permanent address; use it when you cite the number.

Parameters, all optional, in any order:

• income ($ per year), default $210,000 per year.
• dependents (count), default 2.
• increment ($), default $1,000.
• status (one of single, married-joint, married-separate, head-of-household, qualifying-surviving-spouse), default head-of-household.

Values accept plain numbers and loose human formats such as 100k, $100,000, or 6.5%. Unknown parameters are ignored, values outside a field's range are clamped and reported in warnings, and the endpoint never fails on bad input. Responses are cacheable for a day; the defaults change when their sources publish, and tool.version changes when the method does.

• IRS Rev. Proc. 2024-40 (2025 inflation-adjusted items), as of 2025-01-01. The rate tables, standard deduction, and credit amounts for tax year 2025, as published by the IRS.
• One Big Beautiful Bill Act (P.L. 119-21), as of 2025-07-04. The statute that sets the current standard deduction and the $2,200 child tax credit used for 2025.
• 26 U.S.C. (Internal Revenue Code), as of 2025-01-01. The code sections this rate is built out of: § 1 rates, § 63 deduction, § 24 child credit and its phase-out, § 32 earned income credit and its phase-out, § 3101(b)(2) additional Medicare tax.

These are the parameter authorities the engine itself reports for every computation, not citations chosen after the fact. The rate schedule, standard deduction, and credit amounts come from the IRS revenue procedure and the statute that set them; the code sections are the rules whose phase-outs produce the gap this page measures. When the engine loads a different parameter set, this list changes with it.

Carlo is a personal finance agent. It knows your accounts, debts, and goals, so instead of a worked example at the salary you typed it can watch what you actually earn through the year, and tell you what the next raise, bonus, or side contract is really worth before you say yes to it. Text it the question.

ask carlo anything(415) 376-5678