Seven tax myths that scare working families and cost them money

Picture a man at a plant somewhere in the Midwest. He is hypothetical, but you have met him. He is a good worker, the kind who has never missed a shift. His supervisor offers him a few extra weekend hours, and he says no. Why? Because his brother-in-law told him that if you earn one dollar more than the line between two tax brackets, the government takes the higher rate on everything you make, and you end up with less than you started with. So he stays home, mows the lawn, and thinks he has outsmarted the IRS.

He has outsmarted nobody but himself. That belief is false, it has always been false, and it costs real families real money. This is one of those moments where I want to say something that may sting a little, folks: conservatives lose arguments about taxes when we repeat claims that are wrong, and we do it more often than we like to admit. If we want to be the party that trusts ordinary people with their own money, we owe those people the truth about how the money gets taken.

So here are seven claims, stated plainly, tested against the rules, and given a verdict. Most of them circulate on the left and the right alike. One of them is ours. For the arithmetic I will use a hypothetical rate schedule, and I want to be clear that it is made up for teaching purposes. It has three brackets for a single filer: 10 percent on the first $10,000 of taxable income, 12 percent on the next $30,000 (income from $10,000 to $40,000), and 22 percent on everything above $40,000. The real schedule has more brackets and different thresholds, and they change, so when you do your own return use the IRS tables, not mine.

Myth one: a raise can push all your income into a higher bracket

This is the one my hypothetical plant worker believed. The claim says that if your pay crosses a bracket threshold, the new rate applies to every dollar you earn.

It does not. Brackets are marginal. Each rate applies only to the slice of income that falls inside that bracket. Cross a line and only the dollars above the line are taxed at the higher rate.

Work it through with the hypothetical schedule. Say a worker has $39,000 in taxable income and gets a $2,000 raise, bringing it to $41,000.

Before the raise, the tax is:

• 10 percent of $10,000, which is $1,000
• 12 percent of the next $29,000 (from $10,000 up to $39,000), which is $3,480
• Total: $4,480

After the raise, the tax is:

• 10 percent of $10,000, which is $1,000
• 12 percent of $30,000 (all of the second bracket), which is $3,600
• 22 percent of the $1,000 above $40,000, which is $220
• Total: $4,820

The difference is $4,820 minus $4,480, or $340. That is $120 on the $1,000 that stayed in the 12 percent bracket and $220 on the $1,000 that crossed into the 22 percent bracket. Out of a $2,000 raise, $340 went to income tax, and the worker keeps $1,660 of it (before payroll taxes, which are a separate matter). The raise made him richer. It could not have done anything else, because no bracket schedule in American history has taxed the last dollar so hard that it reached back and took the dollars before it.

There are real cliffs in the code, and I will give them their due. Some benefits and credits phase out as income rises, and a family can lose a meaningful amount of help over a narrow band. That is a genuine argument for simplifying those phase-outs. But it is a different argument from "the higher bracket takes everything," and mixing the two up is how people end up turning down overtime. The brackets themselves are not a trap.

Verdict: false. Take the raise. Check any benefit phase-outs separately, in the instructions that come with each form.

Myth two: "I'm in the 22 percent bracket, so I pay 22 percent"

This is the cousin of the first myth, and it sounds like arithmetic. It is not. Your bracket is your marginal rate, the rate on your next dollar. Your effective rate is your total tax divided by your income, and it is always lower once you have more than one bracket.

Take the worker at $41,000. His total tax was $4,820. Divide: $4,820 divided by $41,000 is about 0.1176, or 11.8 percent. He is "in" the 22 percent bracket, and his effective rate on taxable income is under 12 percent. If you hear a politician say that a family in a given bracket "pays 22 percent of their income in taxes," ask which number he means. Often the speaker is using the marginal rate to scare you, or to make a tax cut sound bigger, or to make a tax increase sound smaller. The honest figure for most arguments is the effective rate, and the honest figure for deciding whether to take a second job is the marginal rate. Both numbers matter. They answer different questions.

One more wrinkle. The income that gets taxed is taxable income, which is what is left after the standard deduction or itemized deductions. A family's gross pay and its taxable income can be thousands of dollars apart. Our schedule above is simply applied to the taxable figure. The IRS's Publication 17, Your Federal Income Tax, walks through how gross income becomes taxable income, line by line, and the walk-through in Walking through Form 1040 from the first line to the refund follows the same road.

Verdict: misleading. Know both numbers and use each for its own job.

Myth three: a big refund is a gift from the government

Every spring somebody posts a picture of a refund check as if it were a prize. It is not a prize. It is your own money, returned after the Treasury held it for as long as a year without paying you a dime of interest. I make the full case in your big tax refund is a loan you made to the government for free, so here is the short version.

Withholding is an estimate. Your employer takes money from each paycheck based on the Form W-4 you filled out. If the estimate is too high, you get the overage back after you file. A refund of $3,000 means that over the year you handed the government about $250 a month more than you owed. You were the lender, the Treasury was the borrower, and the interest rate was zero.

Some people like this arrangement because it works as forced saving. Fair enough, and the other essay answers it. But calling it a gift is the myth. If a stranger borrowed $3,000 from you and paid it back a year later with no interest, you would not call the repayment a present.

Verdict: false. It is your money. The only gift in the picture is the one you gave Washington.

Myth four: a deduction and a credit are the same thing

They sound alike. They are not close. A deduction lowers the income you are taxed on. A credit lowers the tax itself, dollar for dollar.

Use the hypothetical schedule. A worker in the 12 percent bracket takes a $1,000 deduction. His taxable income falls by $1,000, so his tax falls by 12 percent of $1,000, which is $120. Now give the same worker a $1,000 credit. His tax falls by $1,000. Same headline number, a result more than eight times larger ($1,000 divided by $120 is about 8.3). And a worker in the 22 percent bracket saves $220 from the deduction, because the deduction is worth more to the person with the higher rate.

That last point is not a footnote. It is the whole reason the choice between deductions and credits is a policy fight, and it is why I wrote four families, one tax change, and how a deduction differs from a credit with four households side by side. A credit is worth the same to every taxpayer who can use it. A deduction is worth more the higher your bracket. Neither design is wicked, but you should know which one a politician is selling you. There is also a second distinction: some credits are refundable, meaning that if the credit exceeds the tax you owe, you receive the difference, and others are not. The instructions for each credit say which kind it is.

Verdict: false. One dollar of credit beats one dollar of deduction for nearly everybody. Read the label.

Myth five: capital gains are taxed just like wages

Here the left and the right both get things wrong, in opposite directions. The left says investors pay nothing. The right sometimes says investors are taxed to death. Neither is right.

The mechanism is this. If you sell an asset such as a stock, a rental house, or a piece of land at a profit, the profit is a capital gain. If you held the asset for one year or less, the gain is short-term and is taxed as ordinary income at your regular rates, the same as wages. If you held it for more than a year, the gain is long-term, and federal law taxes it at lower rates. Long-term gains are taxed at rates of zero, 15 or 20 percent, depending on your income, and a separate 3.8 percent surtax on net investment income reaches higher earners. Qualified dividends get similar treatment. The IRS explains the details in Publication 550, Investment Income and Expenses, and the thresholds are adjusted over time, which is a good reason to look them up rather than trust anyone's memory, including mine.

Why do the lower rates exist? Because part of a long-held gain is only inflation. If you bought a house for $200,000 and sold it for $260,000 after years in which prices generally rose, some of that $60,000 is not real profit, but the tax code taxes the whole nominal figure. A lower rate on gains is a rough way of correcting for that. There is also the argument that taxing income once as wages or profit and again when the saved dollars earn a return discourages saving. Here is the other side's reply: that it is unfair for someone who lives on investment income to pay a lower rate than someone who lives on a paycheck. That is a value claim and a fair one to debate. But the factual claim, that gains and wages are taxed identically, is wrong.

Verdict: false in both directions. Time matters. Hold for over a year and the rate drops.

Myth six: Social Security benefits are never taxed

A great many retirees believe this, and a surprising number discover otherwise at tax time. Social Security benefits can be taxable, depending on what the IRS calls combined income: your adjusted gross income, plus nontaxable interest, plus half of your Social Security benefits. Below the lower threshold, none of your benefits are taxed. In the middle range, up to 50 percent of the benefits may be taxable. Above the upper threshold, up to 85 percent may be taxable. Note the wording: up to 85 percent of the benefit is included in income, which is not the same as an 85 percent tax rate. The included portion is then taxed at your ordinary rates.

The taxation of benefits began with the 1983 Social Security amendments, and the 85 percent tier was added in 1993. Those thresholds were never indexed to inflation, which means that over the years more and more retirees have crossed them. I would rather you check the current figures yourself in IRS Publication 915, Social Security and Equivalent Railroad Retirement Benefits, than take my word for what the numbers are today. The history is worth knowing, though, because it is one of the stealthiest tax increases on the books, and the same principle comes up in How Social Security works and the promise built into its 1935 design and in Inflation is the one tax that never needs a vote in Congress. A threshold that is not indexed catches more people every year, and nobody in Congress has to cast a vote for it.

Verdict: false. Some benefits are taxable, and the share depends on your other income.

Myth seven: tax cuts always pay for themselves

Now for ours. I told you one of these belongs to the right, and this is it. Conservatives, and I count myself among them, like to say that cutting taxes raises revenue. Sometimes in the right circumstances it does. As a blanket rule it does not, and saying it as a blanket rule hands the other side an easy win.

The honest version has three parts. One, lower tax rates can increase the incentive to work, save and invest, and a growing economy produces more taxable income than a shrinking one. That is true, and it is why we argue for cuts. Two, the Laffer curve, which I explain in The Laffer curve explained, and the claim it never actually made, says only that there is a rate so high that cutting it would raise revenue, and that a rate of zero collects nothing. It does not say that every rate is on the wrong side of the peak. Three, when the Congressional Budget Office and the Joint Committee on Taxation score tax bills with "dynamic" assumptions, they generally find that growth offsets some of the revenue lost, but not all of it. That finding is a hard one for the "pays for itself" crowd, and it should not be waved away.

None of this is an argument against cutting taxes. It is an argument for making the real case: that the money belongs to the people who earned it, that lower rates and a broader base grow the economy, and that if spending is the real problem then spending should be cut to match. The 1986 reform that traded lower rates for fewer loopholes, covered in the 1986 tax reform that lowered rates by closing loopholes, is a better model than any slogan.

Verdict: overstated. The growth is real. The free lunch is not.

How to spot the next myth

You do not need a law degree. Ask four questions whenever somebody hands you a tax claim:

1. Is that the marginal rate or the effective rate?
2. Is it a deduction, which lowers income, or a credit, which lowers tax?
3. Who is the "you" in the sentence: a family, a business, or a cherry-picked example?
4. Can I find that rule in an IRS publication or the text of the law?

The fourth question is the one that wins arguments. The IRS publishes its guidance for free, and most of it is more readable than people expect. When the other side makes a claim that fails that test, say so politely. When your own side does, say so too. Credibility is the only asset a political argument has that you cannot print more of.

The bottom line

• Your raise is yours. A higher bracket takes a bigger bite out of the new dollars only, never out of the old ones, so do not turn down overtime on that fear.
• Know your marginal rate and your effective rate, and use the right one for the right decision.
• A refund is your own money coming back without interest. If you want it monthly, adjust your W-4, and use the IRS tools and Publication 505, Tax Withholding and Estimated Tax, before you do.
• Compare credits and deductions dollar for dollar, and read whether a credit is refundable before you count on it.
• Hold investments for more than a year when it makes sense, check Publications 550 and 915 for the current rules, and save the "tax cuts pay for themselves" line for an argument you can actually win.