Each dot represents one household in the middle class in 2018

People on this side will get a tax increase

People on this side will get a tax cut

The tax bill being debated in the Senate this week would affect nearly every American. Numerous analyses have estimated the average impact of the bill on household finances, and advocates on both sides have produced examples of “typical” families that would win or lose under the plan.

Such analyses, however, tend to gloss over the remarkable diversity of Americans’ financial situations. In truth, there is no “typical” American household. Even families that look similar on the surface can differ in ways that radically alter their situation come tax season.

The 25,000 dots on the chart above each represent an American household in the broadly defined middle class. The vertical axis represents income; the horizontal axis represents how big a tax cut (or tax increase) each household would get under the bill in 2018, according to a New York Times analysis using the open-source tax-modeling program TaxBrain. (For details on how we did this analysis, including how we defined the middle class, see the note at the end of this article.)

A couple of things should jump out right away. First, there are more dots on the right side of the chart than the left — more households would get a tax cut than a tax increase. (The chart represents the impact in 2018; the situation looks considerably different in 2027, after many provisions of the bill are set to expire.)

Second, the dots are all over the place. Look at the line of households earning about $100,000 a year. On the far left of the chart, there’s a household that would pay more than $4,000 more in taxes under the Senate plan. On the far right is a household that earns almost the same amount but that would pay $4,000 less under the plan. (All figures are relative to what each household would pay under existing tax law.)

If you want to understand the tax bill, it helps to understand what factors separate those two families, and all the other families in between.

Take these findings with a grain of salt. The tax bill remains a work in progress, and some details aren’t yet clear. Our analysis doesn’t try to account for the impact the bill would have on the economy, for good or ill. (In tax-wonk terms, ours is a “static” analysis rather than a “dynamic” one.) Nor does it account for any spending cuts that Congress might adopt now or in the future to pay for these tax cuts. (The Congressional Budget Office estimates that under congressional budgetary rules, the tax plan would force deep cuts to Medicare spending over the next decade.) But it should give us a sense of how the bill’s many provisions would help — and hurt — millions of families across the country.

Nearly everyone who takes the standard deduction gets a tax cut in 2018

Households that take the standard deduction

Households that itemize

The United States tax code is full of specialized deductions — for charitable donations, mortgage interest, medical expenses and lots more. But about three-quarters of middle-class taxpayers don’t have to worry about any of that — they take the so-called standard deduction, which replaces most of those specialized tax breaks with a single lump sum.

The Senate bill would roughly double the standard deduction: to $12,000 for an individual or $24,000 for married couples. As a result, most middle-class households that take the standard deduction now would get a tax cut under the bill in 2018, and almost none would get a tax increase.

The story is very different for the roughly one-quarter of middle-class families that itemize deductions. The Senate bill would eliminate some popular tax breaks, including deductions for state and local taxes. As a result, households that take those deductions now could lose out. In total, about 40 percent of households that itemize their deductions would pay more in 2018 under the Senate bill — in some cases a lot more.

Families with children generally get a bigger tax cut

Married with no children

Married with one child

Married with two children

The Senate bill would affect different types of families differently. It’s easier to see how the various provisions would play out when we sort the data based on the number of children in each household.

Start with the chart on the left, which shows just married couples with no children. There’s still plenty of variation, but now some clear patterns emerge. For the roughly 70 percent of these couples that take the standard deduction, the G.O.P. bill has three main changes that matter: the doubling of the standard deduction (which helps them), the elimination of the personal exemption (which hurts them) and the revision of the income tax brackets (which would lower tax rates at many income levels).

The heavy diagonal line on the chart reflects the most common situation for these couples. Those earning $40,000 — roughly the bottom of the middle class in our definition — get about a $300 tax cut. Higher earners get a bigger cut, in absolute terms, because they have more income that’s subject to the reduced tax rate. (The fainter vertical line represents mostly older people. Much Social Security income isn’t taxed, so they benefit from the higher standard deduction, but many aren’t affected by the lower tax rate.)

For families with children, another big provision comes into play: the child tax credit. The Senate bill would double that credit to $2,000 per child. As a result, families with children would generally get a bigger tax cut, although the benefits start to phase out above a certain income — that’s why the cuts are smaller for families higher up on the earnings ladder. (This analysis is based on an interpretation of the tax bill that is being used by the Joint Committee on Taxation and many other economists. The bill is ambiguous, however, and there is an alternative interpretation that would be much less generous to lower-income parents.)

People who pay a lot in state and local taxes could see big tax increases

Pay little in state and local taxes

Pay a moderate amount of state and local taxes

Pay a lot in state and local taxes

One of the bill’s most significant changes would be the elimination of the deduction for state and local taxes. More than 40 million households wrote off a combined $350 billion in state and local income and sales taxes in 2015, according to the I.R.S., and 38 million households deducted close to $200 billion in property taxes. Both deductions would disappear under the Senate bill. (The House version of the bill would get rid of the deduction for income and sales taxes but would cap — not eliminate — the property-tax deduction.)

The charts above group households into three broad categories. The group on the left deducts little in state and local taxes; many deduct nothing at all and instead take the standard deduction. The group on the right deducts $4,400 or more in state and local taxes (including income, sales and property taxes), and as a result, many would pay more under the Senate bill.

Many households most affected by the loss of the state and local tax deduction live in coastal states such as New York, California and Connecticut. Those states have high tax rates; they also tend to vote for Democrats. In Connecticut, for example, more than 40 percent of taxpayers deducted either income or sales taxes in 2015, compared with just 28 percent in the United States as a whole.

As the charts show, households that take large state and local deductions also tend to be relatively wealthy. That’s partly because richer households tend to take bigger deductions across the board. But it’s also a reflection of the higher cost of living — and correspondingly higher incomes — in many coastal states.

In 2027 the picture is more uncertain, but many middle-class households would face a tax increase

Excluding impact of corporate tax cuts

Including impact of corporate tax cuts

Until now, we’ve been focusing on the impact of the Senate bill on people’s taxes in 2018, when most households would get at least a small tax cut. But the situation would look very different a decade from now. That’s because in order to reduce the cost of the bill, its authors set essentially all of the individual tax cuts — the doubled standard deduction, the more generous child credit, the lower tax rates — to expire after 2025. But one provision that’s bad for taxpayers — changing the measure of inflation used for many tax calculations — would not expire. As a result, two-thirds of middle-class households would get a tax increase in 2027, and none — zero percent — would get a tax cut. (That’s what’s shown in the left-hand chart above.)

Those figures, however, consider only how the bill would affect personal income taxes. Starting in 2019, the bill would also cut taxes on businesses. Unlike the personal tax provisions, the business tax cuts would not expire.

Exactly how corporate tax cuts would affect individual households is a matter of intense debate among economists. The White House argues that cutting corporate taxes would result in big wage gains for workers. Many independent economists say most of the gains would go to business owners rather than workers. But economists generally agree that companies would probably pass on at least some of their savings to employees. (Executives and other top earners would probably get the biggest boost.) And people who own stocks or have other business investments would see a direct benefit from corporate tax cuts.

Congress’s in-house tax analyst, the Joint Committee on Taxation, estimates that about 25 percent of the cuts to corporate taxes would go to workers in the form of higher incomes. Using that and some other assumptions from the committee, we can allocate the corporate tax cuts to individual households, which is what we show in the chart on the right above. A bit under half of middle-class households would get a tax cut — or, technically, an increase in after-tax income — under these assumptions, and a bit more than half would see their after-tax income go down compared with under existing law. Both the cuts and the increases would generally be small for middle-class households, although the gains could be significant for some wealthy households.

Compared with the 2018 analysis, conclusions about the effects of the tax bill in 2027 are highly uncertain. On the one hand, Republicans argue that Congress is unlikely to allow tax cuts to expire in 2026 as planned, so the middle class might do better than this analysis suggests. On the other, many liberal economists argue that the Joint Committee on Taxation is too generous in its assumptions about how much workers would benefit from the corporate tax cuts. Our analysis makes no effort to account for the broader economic effects of either the tax cuts or of the extra debt that the government would have to take on to pay for them. It also doesn’t account for the repeal of the Affordable Care Act’s individual mandate, a provision that would have a much bigger impact in 2027 than in 2018.

Some details about our sources and methodology

The tax analysis is confined exclusively to middle-class households, defined as households that earn two-thirds to two times household-size adjusted median income. That’s roughly $40,000 to $125,000 for a family of four, or about $30,000 to $90,000 for a couple. For singles, the middle class starts at about $20,000, although to make the charts more readable, we aren’t showing households earning less than $40,000. (We’re using what tax analysts call “expanded income,” which includes cash income but also noncash items such as employer contributions to health insurance and the employer share of payroll taxes.)

To figure out how the Senate bill would affect households, we worked with the Open Source Policy Center — a Washington research organization affiliated with the right-leaning American Enterprise Institute — to model the proposal using the center’s TaxBrain program. Special thanks to Ernie Tedeschi, an economist and occasional Upshot contributor, for his thoughts on modeling the impact of the corporate tax cuts.

The data we’re using comes from the Census Bureau’s Current Population Survey, which asks tens of thousands of Americans detailed questions about their household finances every year. Although these records are not as accurate as the administrative I.R.S. tax records that some think tanks like the Tax Policy Center use (particularly regarding the wealthiest Americans), they line up reasonably closely. The Open Source Policy Center adjusted the data to align with I.R.S. definitions of income; the center also ran parts of our analysis using data from the I.R.S. to ensure that the findings held up.