Take the house out and the median American is 60% poorer

Here is the sentence that made this worth writing. In the 35 to 39 age bracket, the median American household has a net worth of $138,588. Take out the equity in the home they live in and the same median household has $43,416.

That is a drop of 69%, and it is not a rounding adjustment or a methodological quibble. It is the difference between a household that looks comfortably on track and one that has about a year of expenses to its name. Both numbers describe the same people on the same day.

The Federal Reserve's Survey of Consumer Finances, the source almost every net worth article in the US eventually traces back to, publishes both figures. The one including home equity is the one that gets quoted, because it is the headline definition of net worth and because it is the more flattering of the two. The one excluding it sits in the same dataset and, as far as I can find, is published almost nowhere.

The full table

Median household net worth by age of the household head, in nominal 2022 dollars, with and without equity in the primary residence.

Two things stand out. The first is the 18 to 24 row, which barely moves at all, because almost nobody in that bracket owns a home. It is the control case, and it confirms that what the rest of the table is measuring really is housing rather than some artefact of the survey.

The second is that from 30 onwards the drop never falls below half. The worst is the 80+ bracket at 73%. Across every bracket from 25 upwards the average drop is 60%. So for most of adult life, the majority of the median household's net worth is the house they sleep in.

Why this matters more than it sounds

The standard objection is that home equity is real wealth, and that is true. It is an asset, it can be borrowed against, and it is realised in full when the house is eventually sold. Excluding it entirely would be its own distortion.

But there are two situations where counting it produces an answer that is actively misleading, and between them they cover most of the reasons anyone looks up a net worth benchmark in the first place.

The first is comparing yourself to the benchmark when you rent. If you are 37 and renting with $60,000 invested, the standard table says you are well below the $138,588 median and doing badly. The table without housing says you are comfortably above the $43,416 median and doing fine. The second comparison is the honest one, because you are being compared on the same basis rather than against a number that is mostly made of an asset you do not have.

The second is retirement planning. A house you live in does not fund your retirement unless you move out of it, and most people do not. So a 62 year old reading that the median household in their bracket has $392,860 and feeling reassured is reading a number of which roughly 63% cannot be spent without a move they have not planned. The spendable median for that bracket is $143,640, which at a 4 percent withdrawal rate is about $5,746 a year.

That is the number worth sitting with. Not because it is a catastrophe, since Social Security does a great deal of work at that income level, but because it is a very long way from what the headline figure implies.

The other distortion, while we are here

Home equity is not the only reason published net worth figures mislead. The more familiar problem is that a lot of articles quote the average rather than the median, and in this dataset the average is between 2.9 and 4.9 times the median in every single age bracket, because a small number of very wealthy households pull it upwards.

Combine the two and the gap becomes absurd. The average net worth for 40 to 44 year olds is $590,710. The median for the same bracket, with housing stripped out, is $57,668. Both are accurate descriptions of the same group of people. One is roughly 10 times the other.

Whenever you see a net worth figure quoted without both of those qualifications attached, it is worth asking which of the two is doing the work.

Both distortions, in one table

Since the two problems compound, it is worth seeing them side by side. The first column is the number most articles print. The last column is the one that describes the typical household without a house.

Read the 50 to 54 row across. The average is $1,132,497. The median is $266,140. The median without the house is $94,923. Three numbers, one group of people, and the first is about 12 times the last.

Which one is correct depends entirely on the question. If you want to know how much wealth exists in that age group, the average is the right tool, because total wealth is what it measures. If you want to know what a typical household in that group actually has, it is the worst of the three by a distance, and it is the one that gets printed.

A worked example, because the abstraction hides it

Take two households, both 45, both with exactly the same income and the same monthly savings. One bought a house in 2015 and has $220,000 of equity in it plus $90,000 invested. The other has rented the whole time and has $180,000 invested and nothing else.

On the standard measure the owner has $310,000 and the renter has $180,000. The owner looks 72 percent wealthier and sits well above the $213,586 median for their bracket, while the renter sits below it.

On the measure that excludes housing, the renter has twice what the owner has. And if both lost their job next month, the renter is the one who can cover two years of expenses without borrowing or moving. Neither picture is wrong. They are answering different questions, and only one of them gets published.

This is not an argument that renting is better, which depends on rents, house prices and how long you stay. It is an argument that a single number claiming to summarise financial position across households with completely different housing arrangements is doing something dishonest by omission.

What to do with this

If you rent, compare yourself to the third column. That is the like for like comparison and it is almost certainly kinder than the one you have been using.

If you own, track both. The headline number is the one that matters for your estate and for your options later in life. The number without the house is the one that matters for whether you could stop working, handle a bad year, or move without the move being forced.

And in either case the level matters far less than the direction. A benchmark tells you where you sit among people who happen to share your birth year. It says nothing at all about whether your own numbers work, because that depends on what you spend.

Method and sources

All figures come from the Board of Governors of the Federal Reserve System, Survey of Consumer Finances 2022, the most recent edition at the time of writing. Interviews were conducted during 2022 with roughly a quarter completed in 2023, and all values are in nominal 2022 dollars, so they are not inflation adjusted to today.

The unit is the household, or in the survey's own terms the primary economic unit, not the individual. Age refers to the age of the household head. Net worth is total assets minus total liabilities. The second column removes equity in the primary residence, meaning the value of the home minus the mortgage and any home equity borrowing secured against it. It does not remove other property.

One disclosure on provenance, because it matters if you are going to cite this. The percentile anchors underlying these medians were computed from the SCF public microdata by DQYDJ, and cross checked against Wealthtender's republication of the same survey. The arithmetic in this piece, the comparison between the two definitions and the drop percentages, is ours. We are rebuilding the percentile computation directly from the Federal Reserve microdata and this page will be updated with first party figures when that lands, along with the 2025 survey when it is released.

The 18 to 24 bracket rests on a thin sample and its extreme percentiles should be treated with caution. That does not affect its median, which is the only figure used here.

Corrections and questions to usedecoapp@outlook.com. Reuse anything on this page, with or without credit.