Effective Tax Rate

An effective tax rate is the share of income that went to tax, computed after the fact by dividing total tax by a measure of income. It describes a completed return rather than pricing a decision, which is what separates it from the marginal tax rate. The name is worth pausing on, because it is the popular term rather than the official one. No provision of the Internal Revenue Code defines it. The IRS's Statistics of Income division and the Congressional Research Service both call this measure the average tax rate, and CRS reserves the phrase "effective marginal tax rate" for something close to the opposite idea, the tax on the next dollar earned once phase-ins and phase-outs are counted. One CRS report carries all three terms in its title at once. A reader who meets "effective rate" in consumer writing and "average rate" in a government table is looking at the same arithmetic under two names.

The numerator is settled and the denominator is a choice. Everyone agrees the top of the fraction is the tax. What sits underneath it varies by who is computing, and the differences are large enough to matter. IRS Statistics of Income states its method plainly, describing average tax rates as "computed as total income tax divided by adjusted gross income." The Congressional Research Service uses a different base, writing that "the average tax rate is calculated as income tax liability divided by cash income," where cash income is adjusted gross income less state and local tax refunds, plus above-the-line adjustments, tax-exempt interest, and the non-taxable portions of Social Security benefits and pension or annuity distributions. Consumer writing usually divides by gross income or by taxable income. And corporate reporting divides income tax expense by pre-tax book income, which is why a company's reported effective tax rate is not comparable to a household's at all.

Taxable income is the smallest of those denominators, so it produces the largest rate; total income is the largest, so it produces the smallest. The practical rule follows directly. State the denominator in the same sentence as the rate, and treat any effective rate quoted without one as incomplete.

Why the IRS keeps a second income measure. Alongside rates computed on adjusted gross income, Statistics of Income publishes rates computed on what it calls the 1979 Income Concept, which it describes as having been "developed to provide a more uniform measure of income across tax years" by including the same income and deduction items in every year's calculation. The reason is that the statutory definition of adjusted gross income itself changes as Congress adds and repeals adjustments, so a rate computed against it in one year is not strictly comparable to the same rate a decade later. Anyone comparing a personal effective rate against a published average is running into a smaller version of the same problem.

Lower than the bracket rate, usually, and not always. CRS states it with care, noting that "for nearly all taxpayers, average tax rates are less than statutory tax rates." The mechanism is that taxable income moves through the bands in layers while the standard deduction is taxed at nothing, so the top rate reaches only the top slice. Two documented departures are worth knowing, because the usual shorthand does not survive either of them.

The rate can be negative. CRS records that "many taxpayers, particularly lower-income taxpayers, have negative average tax rates," because refundable credits such as the earned income tax credit can pay out more than the income tax owed. A rule stated as "always lower than your bracket" does not describe that household, and the usual explanation for why the effective rate is lower, the standard deduction and the lower bands, does not explain it either.

And a negative income tax rate frequently sits beside a positive payroll tax bill, which is CRS's own observation about the same group of taxpayers. An effective rate built from Form 1040 alone leaves out Social Security and Medicare tax, state income tax, and everything paid at a register or through an escrow account. Build the numerator differently, adding self-employment tax for instance, and the answer can exceed the income tax bracket rate rather than falling below it.

What the number is genuinely good for. Comparing one household's burden against another's, or one year against another, once the denominators match. Sanity-checking a plan, since every deduction and credit eventually shows up in this one ratio. And valuing a change that spans years, such as deciding which years to realize income in. What it cannot do is price a decision. An extra dollar of income, or an extra dollar of deduction, is worth the marginal rate, not the average one.

Two steps and one decision. Total the tax, choose the income measure, then divide. The decision is the part that needs stating out loud, because one return supports several defensible answers.

A hypothetical example. Rosa's total income for the year is $100,000. Above-the-line adjustments of $8,000 bring her adjusted gross income to $92,000, and her deductions bring taxable income to $75,000. Suppose the brackets that apply to her leave a federal income tax of $11,212. Three effective rates follow from that one return. Measured against taxable income, $11,212 divided by $75,000 is 14.9%. Against adjusted gross income, $11,212 divided by $92,000 is 12.2%. Against total income, $11,212 divided by $100,000 is 11.2%.

All three divisions are correct, and they span more than three percentage points. None of them is Rosa's marginal rate, which is the rate charged on the last dollar of that $75,000. If she reads a published table of average rates, the middle figure is the one to compare against, because adjusted gross income is the denominator Statistics of Income uses.

What the measure is good for

• Compresses a whole return into one number, which is what makes it usable for comparing households or comparing one year against another.
• The right basis for a multi-year question, such as which years to realize income or gains in.
• Every deduction, exclusion and credit eventually shows up in this single ratio, so it is a useful check on whether a plan did what it was meant to.

Limits and cautions

• The denominator is not standardized, so two credible sources can report different effective rates for the same return without either being wrong.
• It prices nothing. A deduction or an extra dollar of income is valued at the marginal rate, and using the average rate instead understates both.
• A federal income tax effective rate omits payroll tax, state income tax and consumption taxes, so it understates what a household actually pays.
• Comparing a personal rate against a published average is unreliable unless the income definitions match, which is precisely why the IRS maintains a second, deliberately fixed income measure for its own comparisons.