Retirement Savings Percentile Calculator: Where You Actually Stand

Updated August 2, 2026. Quick answer: enter your age and balance and this ranks you against households the same age, using the Federal Reserve’s own survey data that we computed ourselves. Before you do, the number that reframes the exercise: among households aged 55–64, half have less than $16,600 in retirement accounts, while the average is $306,404. Both are true, and only one describes most people.

Where you stand

The full distribution, by age

Three things this table says that headlines do not

The median peaks lower and earlier than you would think. The highest cohort median is $20,280, among 45–54s. It then falls for 55–64s ($16,600) and again for 65–74s ($5,500) — partly because people spend what they saved, and partly because those with nothing never had it to spend.

Between a third and a half of every cohort has nothing at all. From 37.8% at 45–54 to 58.2% at 75+. Retirement accounts are not a universal feature of American life; they are a feature of having had a job that offered one.

The average is not a typical person. At 55–64 the mean is about 18 times the median. Why that gap exists, and why almost every headline uses the wrong one.

A rank is not a plan

This page answers “where do I stand”, which is the question people ask. It is not the question that matters. Being ahead of most households tells you very little, because most households are not on track either — and being behind tells you less than you fear if your costs in retirement will be low.

The question that matters is whether your number works for your life: the gap between what you need and what you have. If the answer is uncomfortable and you are over 50, the catch-up rules exist precisely for this — what you are allowed to add. And if the number is small but the years are many, what the enhanced catch-up window at 60 to 63 allows is the more useful frame than any percentile.

How these numbers were produced

They are our own computation from the Federal Reserve’s 2022 Survey of Consumer Finances summary extract — 4,595 households — not a figure copied from another article. The method was written down before the computation ran, and the script is deterministic.

The part most summaries get wrong. The SCF is published as five multiply-imputed replicates of every household. Treating the file as one dataset of five times as many households inflates the sample and understates uncertainty. We compute every statistic separately within each of the five, then average — and say so, because a summary that does not mention implicates usually has not handled them.

We checked ourselves against the Board. Against the Federal Reserve Bulletin’s own published table for the same wave: our share of families holding a retirement account matches exactly (54.3% against their 54.3%), and our conditional mean matches to 0.027% ($334,090 against $334,000).

Our conditional median comes to $86,600 against their $86,900 — a 0.345% difference, and we can say exactly where it comes from. It is the weighted-median convention, not the data: two defensible conventions bracket their figure (averaging the five implicates gives $86,600, pooling them gives $87,000), and the five implicate medians themselves span $83,000 to $90,000. A median on this data is genuinely sensitive to how you compute it, which is worth knowing about every published figure of this kind, including ours.

Source: Board of Governors of the Federal Reserve System, Survey of Consumer Finances 2022, Summary Extract Public Data. Variable: RETQLIQ (quasi-liquid retirement assets), whose composition we verified arithmetically against its four components on all 22,975 records. Figures cover all families in each cohort, including those with nothing. Archive SHA-256 recorded in our staged dataset.