Photo by Jorge Campos on Unsplash
The short version
- Average wealth sits well above median wealth because compounded returns are right-skewed — a handful of large outcomes pull the mean up while the typical person stays near the middle.
- The gap is not a data artifact; it is a mathematical property of multiplicative growth. Variance drag (σ²/2) separates the average path from the path most people actually live.
- Bottom line: plan against the median outcome, not the average. Anchoring expectations to the mean quietly overstates what a single, ordinary investor should expect.
Ask what a diversified equity portfolio "returns over the long run" and you will usually get an average — a mean of many outcomes. But no single investor lives the average. They live one path, drawn once, and that path is far more likely to land near the median than the mean. The two numbers diverge, and they diverge in a predictable direction: the average almost always sits above the middle.
This is not a quirk of any one dataset. It is what multiplicative, compounding processes do. Understanding why median wealth trails average wealth is the difference between a realistic retirement plan and one calibrated to an outcome most people will never see.
Context: two numbers that describe the same population
The mean of a distribution is its center of mass — sum every value, divide by the count. The median is the middle value: half the population sits above it, half below. For a symmetric distribution the two coincide. For a right-skewed distribution — a long tail of large values — the mean is dragged upward while the median stays put.
Wealth is emphatically right-skewed. In the Federal Reserve's Survey of Consumer Finances, mean U.S. family net worth has run roughly three to four times the median in recent surveys — the mean reflecting a small number of very large balance sheets, the median describing the household in the middle. When a headline reports "average" household wealth, it is describing a number that comparatively few households possess. The median is the more honest description of the typical experience.
The same asymmetry appears inside a single investor's return distribution over time, and that is where it matters most for planning.
Why compounding manufactures skew
Returns compound multiplicatively, not additively. A portfolio that gains 50% and then loses 50% is not flat — it is down 25%, because the loss applies to a larger base. Chain many such multiplicative steps together and the distribution of terminal wealth becomes approximately log-normal: bounded at zero on the downside, unbounded and long-tailed on the upside. The logarithm of wealth is roughly symmetric; wealth itself is skewed.
That skew is why the average path and the median path separate. The arithmetic mean of annual returns describes the expected value of wealth — a number pulled upward by the rare, enormous upside sequences. The geometric mean describes the growth rate of the median path, the compounded rate an ordinary sequence actually earns. The distance between them is set by volatility:
The average is the outcome of the luckiest sequences averaged in; the median is the outcome you should actually plan to receive.
The approximation is compact. Geometric return ≈ arithmetic return − σ²/2, where σ is the standard deviation of returns. That σ²/2 term is variance drag — the same mechanism that quietly erodes leveraged and high-volatility strategies. It is not a fee, a tax, or a behavioral error. It is arithmetic. Higher volatility widens the gap between the average an investor is quoted and the median they experience.
Working the numbers
Take a plausible broad-equity assumption: an arithmetic mean annual return of 8% with a standard deviation of 16%. These are round illustrative figures, not a forecast. The variance drag is 0.16² / 2 = 0.0128, or about 1.3 percentage points. So the geometric — the median-path — growth rate is roughly 8% − 1.3% = 6.7%.
Project one dollar forward 30 years under log-normal assumptions:
- Median terminal wealth grows at the geometric rate: 1.06730 ≈ 7.0×.
- Mean (expected) terminal wealth grows at the arithmetic rate: 1.0830 ≈ 10.1×.
Same assumptions, same 30 years — and the average outcome is roughly 44% larger than the median. An investor who budgets retirement around "10×" is planning against a number that fewer than half of otherwise-identical portfolios will reach. The median investor lands closer to 7×. And because the drag scales with σ², a more volatile allocation does not merely add risk; it widens the mean–median gap, making the average an even worse proxy for the typical result.
Initially I treated the mean as the natural planning anchor — it is, after all, the "expected value." Then I sat with the fact that expected value is an average over parallel universes I will never occupy. I get one draw. For a single draw from a right-skewed distribution, the median is the more defensible planning number, and the mean is a description of the tail's gravitational pull.
Why this compounds into behavior, not just spreadsheets
The mean–median gap is not an abstraction that stays on the page. It shapes three concrete decisions.
Expectations and disappointment. Marketing and casual conversation quote averages. If an investor internalizes the average as the baseline, they will experience an ordinary — perfectly successful — median outcome as a shortfall. That perceived shortfall is exactly the condition under which people abandon a disciplined plan late, locking in the gap rather than riding through it.
Sequence risk sharpens the point near the finish line. The median-path math assumes returns arrive in a benign order. They do not always. Sequence-of-returns risk means the same set of annual returns, reordered, can produce materially different terminal wealth once withdrawals begin. Skew and sequence interact: the unlucky orderings cluster in the left tail that the median already sits above, and drawdowns there take longer to repair — which is why I track drawdown recovery time, not just maximum drawdown.
Volatility is a first-order cost, not just a comfort question. Because the gap is σ²/2, reducing portfolio variance does more than smooth the ride — it moves the median path closer to the arithmetic mean. Rebalancing discipline, diversification, and cash buffers are not only about sleeping well; they are about narrowing the distance between the return you are quoted and the return you keep. Small, faithful reductions in variance compound in the investor's favor over decades.
What the macro backdrop does and doesn't change
None of this is regime-dependent in its structure, but the inputs move. As of mid-2026 the 10-year Treasury yields 4.58% and the effective federal funds rate sits at 3.63% (FRED, asof 2026-07-14 and 2026-06-01). With CPI running about 3.7% year over year (FRED, asof 2026-06-01), the real risk-free rate is modestly positive — a meaningfully different starting point than the zero-rate decade. Equity-market volatility, proxied by the VIX at 16.5 (FRED, asof 2026-07-14), is subdued relative to its long-run average.
A calmer volatility regime narrows variance drag in the near term; a higher risk-free rate raises the bar equities must clear to justify their skew. Neither repeals the mean–median gap. They only shift the numbers plugged into σ²/2. The lesson is structural: whatever the regime, plan against the middle of the distribution and treat the average as what it is — a summary weighted by outcomes you are unlikely to draw.
FAQ
Is the mean "wrong" and the median "right"? Neither is wrong; they answer different questions. The mean is the correct expected value across all possible paths — useful for pricing and for aggregating a whole population. The median is the better description of what one investor, on one path, is most likely to receive. For personal planning, the median is the more conservative and more realistic anchor.
Does diversification eliminate the gap? No, but it shrinks it. Because the gap equals σ²/2, lowering portfolio volatility through diversification and rebalancing moves the median path toward the arithmetic mean. It cannot close the gap entirely as long as returns are volatile and compound multiplicatively.
Why does the average keep rising even when most people feel behind? Right-skew. A small number of very large outcomes lift the mean without lifting the median. In wealth data this shows up as a mean that runs several times the median; in return data it shows up as expected terminal wealth exceeding the typical terminal wealth. Feeling "behind the average" is the normal condition for anyone at or below the middle — which is, by definition, half of everyone.
How large is the gap in practice? It scales with the square of volatility and with time. At 16% annual volatility the drag is about 1.3 points a year; over 30 years that compounds into a mean roughly 40–45% above the median in the illustrative case above. Double the volatility and the annual drag roughly quadruples, widening the gap considerably.
Does this argue against holding equities? Not at all. Equities' long-run compounded (geometric) return has historically rewarded patient holders. The point is calibration: use the geometric, median-path number when you set expectations and withdrawal plans, and recognize that reducing unnecessary volatility improves the outcome you are statistically most likely to receive.
Key takeaways
- Median wealth trails average wealth because compounded returns are right-skewed — a mathematical property of multiplicative growth, not a data flaw.
- The gap between the average path and the typical path is variance drag, σ²/2. It grows with volatility and with time.
- For a single investor drawing one path, plan against the geometric (median) return, not the arithmetic (mean) return.
- Lowering portfolio variance through diversification and rebalancing discipline narrows the gap — a first-order benefit, not just a comfort one.
- Regime shifts move the inputs (volatility, real rates) but never repeal the structure. Calibrate expectations to the middle of the distribution.
Editor's read
The most useful thing an investor can do with this idea is unglamorous: quote themselves the geometric number. When the editor stress-tests a long-horizon plan, the median path — not the expected value — is the anchor, because it is the outcome a single life is most likely to deliver. That reframing does two things at once. It sets expectations that a normal, successful decade won't feel like failure, and it turns volatility reduction into a measurable return improvement rather than a matter of nerves. Faithfulness in small things — a few basis points of variance saved, held across decades — is where the median quietly catches up to the mean.
Editor's holdings disclosure: this article discusses no specific security; the editor holds a diversified long-term equity and bond allocation and no single-name position relevant to the analysis.
Methodology. Illustrative figures use an assumed arithmetic mean return of 8% and annual standard deviation of 16% under a log-normal model; geometric return approximated as arithmetic − σ²/2, terminal-wealth multiples computed over a 30-year horizon. These are teaching assumptions, not forecasts. Wealth distribution characterization references the Federal Reserve Survey of Consumer Finances (mean-to-median ratio, recent surveys). Macro figures from FRED: 10-year Treasury 4.58% (asof 2026-07-14), effective federal funds rate 3.63% (asof 2026-06-01), CPI year-over-year ~3.7% (asof 2026-06-01), VIX 16.5 (asof 2026-07-14). Data pulled 2026-07-16.
This article is for educational purposes and does not constitute personalized financial advice. See our full Disclaimer.