The tax-bracket myth: why a raise never lowers your take-home pay

A coworker once told me she turned down a $4,000 raise because it would "push her into the next bracket" and she'd end up with less money. She was wrong, but not stupid — the U.S. tax code is deliberately opaque about a mechanic that would take ninety seconds to teach in a high school classroom. Let's teach it here.

How brackets actually work

The United States uses a marginal tax system. That word — marginal — is doing all the work, and almost nobody explains it. It means the tax rate only applies to the dollars inside that bracket, not to your entire income.

Here are the 2024 federal brackets for a single filer, rounded for clarity:

• 10% on income up to $11,600
• 12% from $11,600 to $47,150
• 22% from $47,150 to $100,525
• 24% from $100,525 to $191,950
• 32%, 35%, and 37% at higher tiers

Now imagine you earn $50,000. A shocking number of people believe this means they pay 22% on the whole $50,000 — $11,000 in federal tax. That is not how it works. What actually happens:

• The first $11,600 is taxed at 10% → $1,160
• The next $35,550 (from $11,600 to $47,150) is taxed at 12% → $4,266
• The last $2,850 (from $47,150 to $50,000) is taxed at 22% → $627

Total federal tax: $6,053. That's an effective rate of about 12.1%, not 22%. Your marginal rate is 22% — the rate on your next dollar earned — but your effective rate is what you actually paid across your whole income. These are two different numbers and confusing them is the entire source of the myth.

So what happens if you get a $5,000 raise and cross deeper into the 22% bracket? You pay 22 cents in federal tax on each of those new dollars. You keep 78 cents. You do not retroactively pay 22% on the $47,150 below the threshold. That money is grandfathered into its own bracket forever. A raise, mathematically, cannot reduce your take-home pay. Full stop.

Why the myth won't die

Three reasons, and they're all structural.

First, withholding is not the same as taxation. Your employer's payroll system estimates what to withhold based on assumptions about your annual income. When you get a mid-year raise, the system sometimes over-corrects for a paycheck or two, assuming you've been earning at the new rate all year. Your paycheck can genuinely look smaller than expected for a cycle, which feels like proof of the bracket myth. It isn't — you'll get the overwithheld money back at tax time, or you can adjust your W-4 to fix it immediately.

Second, benefits cliffs are real, and people conflate them with tax brackets. Subsidies like the ACA premium tax credit, SNAP, Medicaid, and childcare assistance can drop off abruptly at specific income thresholds. A $2,000 raise that costs you a $6,000 healthcare subsidy is a genuine problem — but it's a benefits cliff, not a tax bracket. These are worth calculating carefully, especially if you're near an ACA subsidy edge or receive housing assistance. Tax brackets are gradual; benefits cliffs are not.

Third — and this is the one nobody says out loud — telling employees that raises are barely worth it is convenient for employers. If you believe the myth, you're less likely to push hard for a bigger number. The reason the myth survives in workplaces is that it costs your boss nothing and saves them money.

What to do with this

Stop turning down money. Stop letting your uncle at Thanksgiving tell you a promotion isn't "worth it because of taxes." And when you're negotiating a raise, do the actual arithmetic on what you'll keep, not what you're afraid of losing.

Here's the calculation, roughly. If you're a single filer moving from $70,000 to $80,000, the entire $10,000 raise sits in the 22% federal bracket. Add roughly 7.65% for FICA (Social Security and Medicare) and, say, 5% for a middle-of-the-road state income tax. You're looking at about 34.65% total, meaning you keep around $6,535 of that $10,000. That is not "barely anything." That is $6,535 you did not have before, every year, compounding into raises calculated as percentages of your new base.

When your manager or HR floats a smaller number than you expected and hints that taxes will eat it anyway, you can say:

"I appreciate the offer, but I want to be clear that I'm evaluating this on the pre-tax number, not the after-tax number — the tax rate is the same whether the raise is $3,000 or $8,000, so it doesn't change how I'm thinking about the gap between what's on the table and what I asked for. Can we talk about closing that gap?"

This does two things. It signals that you understand marginal rates, which quietly disqualifies the bracket handwave. And it drags the conversation back to the actual number, which is where you want it.

The one place the myth is almost true

There is exactly one scenario where a raise can hurt you at the federal level, and it's narrow: if you're on the edge of losing a specific tax credit (not a deduction) that phases out abruptly. The Saver's Credit, for example, drops in stair-step fashion — you can lose hundreds of dollars in credit by earning one extra dollar of income. The Earned Income Tax Credit phases out gradually but can still create weird incentives near its boundary.

These are edge cases, and they mostly affect lower- and middle-income filers with specific credits in play. If you suspect you're near one of these thresholds, run your numbers through a free calculator (the IRS Withholding Estimator is fine, or a tool like SmartAsset's tax calculator) before accepting or refusing a raise. But for the overwhelming majority of workers, the bracket myth is just that — a myth.

Pull up your most recent pay stub right now and look at the line labeled "federal income tax" or "FIT." Divide that number by your gross pay for the same period. That's your effective federal rate. If it's lower than the bracket you thought you were "in," you already have your proof.