I have heard this used in arguments (and even used it myself several times): that 20% of the US population pays 80% of the taxes. But I am starting to wonder if that is statistically accurate.

Also, as a sub-question, I have also heard that 40% of people in the US don't pay taxes at all. These are specifically for US Federal income tax.

Are either of these true, or are they just myths?

  • 1
    The devil is in the details. Are you just talking about federal income tax or all taxes at the state and federal level combined?
    – JohnFx
    Commented Apr 1, 2012 at 15:47
  • 7
    This is known as the Pareto principle also called 80-20 rule. It's not "statistically accurate" in the sense that that this numbers are the result of an actual statistic, but an often made observation that roughly 80% of the effects come from 20% of the causes. Also, about 80% of a companies turnover comes from about 20% of customers (the bigger customers). Probably 80% of the question here are from about 20% of the users, etc. It doesn't have to be always 80-20, but the tendency is usually true. Commented Apr 1, 2012 at 17:11
  • 3
    Please remember that "income tax" does not include "payroll tax." The former is predominantly paid by the rich, but the latter only applies to the first 100K of income and so is very regressive. You should always be a bit careful when thinking about taxes to remember that "payroll tax" is not considered income tax even though most people don't think of them as different. Commented Apr 2, 2012 at 1:35
  • 2
    @NoahSnyder: "payroll tax" is social and medical insurance premium, obviously is regressive, as there is limit to how much premium you'd need to pay. Medical treatments cost same regardless of if you're rich or poor.
    – vartec
    Commented Apr 2, 2012 at 19:36
  • 1
    @vartec Another thing people tend to forget when considering the cap on SS tax is that the benefit payouts are also capped when you retire. It isn't a giveaway to rich people as some would have you believe.
    – JohnFx
    Commented Dec 5, 2012 at 23:29

2 Answers 2


Yes, it's pretty much true if by taxes you mean federal Income Tax. Both in case of 20% paying 80% of taxes, and bottom 40% paying insignificant amount of taxes.

Data from 2009:

  • Top 1% (income above $343,927) paid 36.73%
  • Top 5% (income above $154,643) paid 58.66%
  • Top 10% (income above $112,124) paid 70.47%
  • Top 25% (income above $66,193) paid 87.30%
  • Top 50% (income above $32,396) paid 97.75%
  • Bottom 50% (income below $32,396) paid 2.25%

Source IRS via National Taxpayers Union

  • 3
    These numbers do not include Payroll taxes(FICA). For that reason I hate this statistic because it makes it look like the lower income levels are contributing less than they actually are. I think for this statistic to really be reflective of federal taxes being paid all taxes need to be considered.
    – Chad
    Commented Apr 2, 2012 at 14:37
  • 10
    @Chad: in case of FICA taxes, you're contributing to your Social Security, and more you contribute, more you get. It's an insurance premium, and doesn't qualify for strict definition of tax.
    – vartec
    Commented Apr 2, 2012 at 14:47
  • 9
    But the SS Fund has been being used to fund the government. It is not a savings or retirement account. It is a tax collected on payroll. It is not insurance. It is a federal program that has expenses that is managed by the government. It is not optional and I can not opt out of it. If it looks like a tax (Takes money from me base on my pay and gives it to the government), Acts like a tax (Fund federal government operations), its a tax!
    – Chad
    Commented Apr 2, 2012 at 14:52
  • 2
    SS is not a constitutionally mandated entitlement. The US does not have any of those. There is nothing in the constitution that would cause payroll taxes to cease should congress do away with the entitlements that FICA is supposed to fund.
    – Chad
    Commented Apr 2, 2012 at 15:28
  • 5
    Would be a lot more interesting to see breakdown by total tax burden. There are so many taxes, local, regional and national, that it’s hard to know just how much of one’s income goes right back to a branch of the government...
    – RomanSt
    Commented Apr 2, 2012 at 18:46

For Federal Income tax you can get the actual numbers from the IRS Tax stats webpage. There are a few complications to figuring this out including:

  • The IRS numbers are broken down by Income into ranges, so getting exactly 20% is tricky.
  • These numbers only include people who filed a tax return, people with no income are not included in these numbers, but I assume you meant to include them in your 20% number
  • A lot of people file jointly, these numbers are based on the # of returns, not people.

What I was able to calculate in my trusty Excel spreadsheet that is close to an answer is this:

Top 16.7% by adjusted gross income ($100K+) pay 69% of the federal income taxes.
The top 27.7% by adjusted gross income ($75K+) pay 79.5% of federal income taxes.

So to get technical the claim that 20% pay 80% is false since the top 27% don't pay quite that much. That said, the numbers aren't so far off that the underlying point doesn't have merit.

The numbers get more lopsided if you look at the tippy top of income earners with the top 1% paying almost 27.6% of all federal income taxes.

Note: These are 2009 numbers. They, of course, vary by year.

  • 4
    You shouldn't need to do any calculations using the numbers from the IRS spreadsheets to answer the question. Looking at the table for "All Returns: Selected Income and Tax Items" classified by size of adjusted gross income (irs.gov/pub/irs-soi/09in11si.xls), you see that taxpayers with income over $75k comprise the top 20.6% (cell 60C). You then see that this same group of taxpayers pays 84% of all income tax (cell 60R). This is straight from the IRS, without doing any other calculations. Commented Apr 17, 2012 at 16:53
  • 2
    Looking at that same table I also noticed that you can also see that the top 0.5% (cell 57C) pays 29.8% (!!) of all income tax (cell 57R). That is very lopsided, indeed. (We can't get the data for the top 1%, because the groupings jump from 0.5% to 2.8%). Commented Apr 17, 2012 at 16:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .