Introduction: what a net worth percentile can and cannot tell you
A net worth percentile answers a very human question — how far ahead, near the middle, or behind am I compared with people my age — and the answer is only as good as two things: whose distribution you are being compared against, and how that distribution is shaped. Most quick percentile tools get the first roughly right and the second badly wrong, because they draw a straight line between "median" and "rich" and wealth is nothing like a straight line.
This page compares you against United States families in the same age band, using the Federal Reserve's 2022 Survey of Consumer Finances — the source that produces the medians everyone quotes. It uses two published statistics per age band, the median and the mean, and fits a lognormal distribution that reproduces both exactly. That matters because the gap between median and mean is what encodes the skew: for families headed by someone aged 35 to 44 the median net worth is $135,600 while the mean is $549,600, a ratio of four to one, and any model that ignores that ratio will put ordinary balance sheets in the top few percent.
The word to keep hold of is estimated. This is a two-parameter fit to two published numbers, not survey microdata, and the section on limitations below is specific about where it goes wrong. It is designed for the question "is my balance sheet roughly typical, clearly ahead, or clearly behind for my age", which it can answer, rather than "am I exactly at the 87th percentile", which no two-parameter model can.
How to use the net worth percentile calculator
- Pick your age band from the six the SCF publishes. Not ten, not five-year steps — six, because those are the bands the Federal Reserve reports medians and means for, and inventing narrower bands would be inventing precision the source does not have.
- Enter net worth, not assets. Everything you own minus everything you owe: mortgage balance, student loans, car loans and credit card balances all come off. This is the single most common input error, and it moves the answer more than any other.
- Include home equity and retirement accounts. The SCF definition of family net worth includes both, so excluding them compares your number against a benchmark that includes other people's. The asset checkboxes are a reminder of what to add up, not a filter applied to the maths.
- Read the percentile table, not just the headline. The result lists what the 10th, 25th, 50th, 75th, 90th, 95th and 99th percentiles are worth in dollars for your age band. Those cut-points are far more useful than your own single number, because they show how much of the distribution sits between any two rungs.
- Treat household income as context. The percentile depends on age band and net worth only. Income produces the net-worth-to-income ratio, and the education and employment fields appear in the downloadable report; none of the three enters the ranking, because the SCF statistics this model is calibrated against are published by age.
The formula: fitting a lognormal to a published median and mean
Wealth within an age band is approximately lognormal — the logarithm of net worth is roughly normally distributed — which is why a straight-line model fails so badly in the upper half. A lognormal has two parameters, and two published statistics pin both. Writing for the median and for the mean, the log-scale parameters are:
because a lognormal's median is and its mean is . Your percentile is then the normal cumulative distribution of your standardised log net worth:
and the inverse, which produces the dollar value of any percentile in your band, is:
Plain-text formula: sigma = sqrt(2 * ln(mean / median)); percentile = 100 * normalCdf(ln(netWorth / median) / sigma); dollarsAtPercentile = median * exp(sigma * zScore), with the median and mean taken from the SCF 2022 table for the selected age band.
The construction has one property worth stating plainly: it reproduces the published median and mean of every age band exactly, by design. Entering the SCF median for your band returns the 50th percentile to the decimal. That is a consistency guarantee, not evidence that the shape between and beyond those two anchors is right.
Why the old straight-line approach failed
The previous version of this page scaled linearly, treating four times the median as the top of the scale. For a 35-to-44 household that put anyone above roughly $542,000 at the 99th percentile. The lognormal fit puts $500,000 at about the 78th percentile for the same band, and $1,000,000 at about the 88th. The difference is not a rounding adjustment; it is the difference between telling a comfortable saver they are in the top 1% and telling them they are in the top quarter.
Worked example: a 35-to-44 household with $500,000
The SCF 2022 figures for families headed by someone aged 35 to 44 are a median net worth of $135,600 and a mean of $549,600.
- The mean-to-median ratio is 549.6 ÷ 135.6 = 4.053, so σ = √(2 × ln 4.053) = √2.799 = 1.673.
- Standardising $500,000: ln(500,000 ÷ 135,600) = ln 3.687 = 1.305, divided by 1.673 = z = 0.780.
- Φ(0.780) = 0.782, so the estimated percentile is 78th — ahead of roughly four in five families in that age band, and nowhere near the top 1%.
Running the inverse over the same band gives the whole ladder, which is the more useful output:
Estimated net worth at each percentile, families headed by someone aged 35 to 44 (SCF 2022 anchors, lognormal fit)
| Percentile |
z-score |
Estimated net worth |
| 10th | −1.282 | $15,890 |
|---|
| 25th | −0.674 | $43,872 |
|---|
| 50th (median) | 0.000 | $135,600 |
|---|
| 75th | +0.674 | $419,112 |
|---|
| 90th | +1.282 | $1,157,199 |
|---|
| 95th | +1.645 | $2,125,083 |
|---|
| 99th | +2.326 | $6,645,617 |
|---|
Two things stand out. The rungs are multiplicative rather than evenly spaced — going from the 50th to the 75th percentile takes roughly three times the money, and from the 75th to the 90th takes nearly another three — which is exactly the property a linear model destroys. And the same $500,000 balance sheet that ranks 78th for a 35-to-44 household ranks about the 57th for a 55-to-64 household, because the benchmark it is measured against has more than doubled. Percentile is relative to peers, not to a dollar amount.
The underlying data
2022 Survey of Consumer Finances: family net worth by age of head, in 2022 dollars, with the fitted dispersion
| Age of head |
Median net worth |
Mean net worth |
Mean ÷ median |
Fitted σ |
| Under 35 | $39,000 | $183,500 | 4.71 | 1.760 |
|---|
| 35–44 | $135,600 | $549,600 | 4.05 | 1.673 |
|---|
| 45–54 | $247,200 | $975,800 | 3.95 | 1.657 |
|---|
| 55–64 | $364,500 | $1,566,900 | 4.30 | 1.708 |
|---|
| 65–74 | $409,900 | $1,794,600 | 4.38 | 1.719 |
|---|
| 75 or more | $335,600 | $1,624,100 | 4.84 | 1.776 |
|---|
Median net worth rises through working life, peaks in the 65–74 band and falls slightly after 75 as retired households spend down. The mean-to-median ratio is remarkably stable at roughly four across every band, which is why the fitted dispersion barely moves: wealth inequality within an age group is about as severe at 30 as it is at 70.
Limitations you should read before quoting a percentile
The lognormal understates the extreme top. Real wealth distributions have a Pareto tail that is fatter than any lognormal. The fit above puts the 99th percentile for a 35-to-44 household at $6.65 million; the true figure is higher. Above roughly the 95th percentile, treat the output as "very high" rather than as a number.
Zero and negative net worth cannot be ranked by this model. A lognormal is only defined above zero, and a meaningful share of US families — particularly under 35 — owe more than they own. The calculator accepts a negative figure and says plainly that it sits in the bottom of the distribution without inventing a percentile for it.
The data is a 2022 snapshot in 2022 dollars. The SCF runs every three years. House prices, equity markets and the price level have all moved since, so a 2026 balance sheet is being compared against a 2022 benchmark. The comparison is directionally useful and the level is stale in a direction that flatters recent asset owners.
It is national and by age only. No adjustment for region, household size, marital status, homeownership, race, or whether a household is one earner or two — all of which the SCF shows to be large. A $400,000 net worth means something different in San Francisco than in rural Ohio, and this model cannot see the difference.
Age of head is not your age. The SCF classifies a family by the age of its reference person. A couple with a large age gap sits in one band, not two.
Income, education and employment do not affect the ranking. They are collected for the net-worth-to-income ratio and the downloadable report. Making them move the percentile would require conditional distributions the published tables do not supply, and inventing them would be worse than leaving them out.
This is not financial advice. A percentile is a comparison, not a plan. Being above the median for your age says nothing about whether you are on track for your own retirement, which depends on your spending, your timeline and your obligations rather than on where other people stand.
Questions about net worth percentiles
Where do the benchmark numbers come from?
From the Federal Reserve Board's Survey of Consumer Finances for 2022, published in October 2023, which reports family net worth by age of head in 2022 dollars. The calculator uses both the median and the mean for each of the six published age bands, because the ratio between them is what encodes how skewed the distribution is. Using the median alone, as most quick tools do, throws that information away.
Why does this give a much lower percentile than other calculators?
Because most of them scale linearly from the median, which treats a few multiples of the median as the top of the distribution. Wealth is roughly lognormal, so the rungs of the ladder are multiplicative: for a household aged 35 to 44 the 75th percentile is about three times the median and the 90th is about three times that again. A linear model puts a half-million-dollar balance sheet in that age band near the 99th percentile; the lognormal fit puts it around the 78th.
What exactly counts as net worth?
Everything you own minus everything you owe. Assets include cash, checking and savings, retirement accounts, taxable investments, the market value of a home and other real estate, business interests and vehicles. Liabilities include the mortgage, home equity loans, student debt, car loans, credit card balances and any other debt. Because the survey definition includes home equity and retirement accounts, leaving either out compares your figure against a benchmark that includes other people's.
What if my net worth is zero or negative?
The model cannot rank it, and says so rather than guessing. A lognormal distribution is only defined for positive values, and a meaningful share of US families, particularly under 35, have negative net worth from student loans, a new mortgage or business debt. The calculator will confirm you are in the lower part of the distribution and skip the numeric percentile instead of printing a made-up figure.
Does my income change my percentile here?
No. The ranking is a function of your age band and your net worth only, because the published statistics it is calibrated against are reported by age. Household income is used for the net-worth-to-income ratio, which is a useful context check on whether a balance sheet is large relative to the earnings that built it, and education and employment appear in the downloadable report. Making any of them move the percentile would require conditional distributions the published tables do not provide.
Sources. All benchmark figures are from the Board of Governors of the Federal Reserve System, "Changes in U.S. Family Finances from 2019 to 2022: Evidence from the Survey of Consumer Finances", published October 2023. Family net worth by age of head, in 2022 dollars, median and mean respectively: under 35, $39,000 and $183,500; 35–44, $135,600 and $549,600; 45–54, $247,200 and $975,800; 55–64, $364,500 and $1,566,900; 65–74, $409,900 and $1,794,600; 75 or more, $335,600 and $1,624,100. The percentile is produced by fitting a two-parameter lognormal that reproduces both of those statistics exactly for the selected band; the distributional shape between and beyond the anchors is an assumption of this page, not a Federal Reserve result, and it understates the extreme upper tail because real wealth distributions are Pareto rather than lognormal at the top. The SCF is conducted every three years, so 2022 is the most recent published wave and its figures are not adjusted here for subsequent asset price or price level changes. This page performs arithmetic on the figures you enter, stores nothing, transmits nothing, and is not financial advice.