Over the long windows measured to August 2026, the S&P 500 has returned more than gold after inflation in five of seven cases. Since January 2000, gold has returned more. Both numbers come from the same data, and the starting month decides which one gets quoted. That is why most versions of this comparison read as arguments rather than measurements. The more durable finding sits elsewhere: month to month, the two assets have almost no relationship at all.
Key takeaways
- Five of the seven fixed windows end with the S&P 500 ahead on real total return. The window that starts in January 2000 ends with gold ahead, 8.1% against 5.7% a year after inflation.
- Comparisons that set gold’s price against the S&P 500 price index leave out dividends and understate stocks in every window.
- Gold has not been the calmer holding: annualized volatility of monthly returns from 1960 to 2026 runs 15.2% for gold against 12.5% for the S&P 500’s real total return.
- The correlation between the two series’ monthly returns is 0.00 over the full record. They have not moved together.
Why most versions of this comparison are wrong
The method decides the answer, and two errors are common.
The first error compares gold’s price with the S&P 500 price index. The price index tracks share prices alone. Dividends are part of a stock investor’s return, and over long windows they are a large part of it. A comparison that leaves them out understates stocks. The correct comparison uses total return: price change plus reinvested dividends.
The second error quotes a single window. One starting month can produce either answer. A window that opens at a gold peak makes gold look bad, and a window that opens at a stock peak does the reverse. An honest version shows several windows and names the starting month for each.
This page uses fixed windows, total return, inflation-adjusted figures, and one named source per series.
The record, window by window
Stocks lead in five of the seven windows below; gold leads in the one that begins in January 2000. The figures come from three public series, processed the same way.
Gold prices are monthly averages from the World Bank Commodity Price Data (“Pink Sheet”), downloaded September 22, 2026. Inflation comes from the Consumer Price Index for All Urban Consumers (CPI-U, series CPIAUCSL), published by the U.S. Bureau of Labor Statistics and served by the Federal Reserve Bank of St. Louis, downloaded September 23, 2026. Stock returns come from Robert Shiller’s U.S. stock market dataset, downloaded September 23, 2026. Shiller’s real total return series reinvests dividends and is already adjusted for inflation. Gold figures are restated in August 2026 dollars with the same index. Every figure is a monthly average, never an intraday price. Growth rates are compound annual rates to August 2026. The stock figures include reinvested dividends. The gold figures cannot, because gold pays none.
Real compound annual growth to August 2026:
| Bought in | Gold | S&P 500, total return |
|---|---|---|
| August 2016 | 9.0% | 11.6% |
| August 2006 | 7.5% | 8.7% |
| January 2000 | 8.1% | 5.7% |
| August 1996 | 5.8% | 7.7% |
| August 1976 | 3.9% | 8.0% |
| January 1970 | 4.8% | 6.9% |
| January 1980 | 0.9% | 8.8% |
The widest stock lead sits in the window that begins at the January 1980 gold peak: 8.8% a year for the index against gold’s 0.9%. Anyone can pick a window that proves either case, which is why the window has to be stated whenever a figure is.
Volatility
Gold has been the more volatile of the two, not the less. Annualized volatility of monthly returns from 1960 to 2026 is 15.2% for gold and 12.5% for the S&P 500’s real total return. Over the last ten years the two are close: 12.1% for gold and 12.5% for the index. The claim that gold is the calmer holding does not survive the full record. Across it, gold has moved more, not less. Volatility measures the size of the swings, not their direction.
The falls
Neither record is a smooth line, and the recovery times differ by decades. Gold fell 62.1% in price terms from January 1980 to July 1999. After inflation the same fall reaches 83.0%, measured from January 1980 to April 2001, and gold did not match January 1980’s inflation-adjusted price until February 2025. The S&P 500’s real total return fell 51.8% from August 2000 to March 2009, measured the same inflation-adjusted way. A single starting month inside either fall produces the grim figures that single-window comparisons quote. Gold’s full inflation-adjusted record is set out in is gold a good investment?.
Where the two actually differ: they do not move together
Gold and the S&P 500 have not moved together. Correlation runs from −1, for two series that always move in opposite directions, to +1, for two that always move in step. Measured on monthly returns, the correlation between gold and the S&P 500’s total return is 0.00 over the full series and 0.10 over the last ten years. Both figures sit at the no-relationship end of the scale. Gold and silver, measured the same way, stand at 0.69 over the full series and 0.73 over the last ten years. Two holdings with no month-to-month relationship behave differently in the same conditions. That is a separate question from which one returned more, and no allocation follows from the correlation alone. The two metals are compared in gold vs silver.
The S&P 500 priced in gold
The ratio behind the charts divides the S&P 500’s level by the gold price. The result is the number of ounces of gold that one unit of the index costs. At the August 2026 monthly averages, the index stood at 7,711 and gold at $4,411, so the index cost 1.75 ounces. The series high was 5.42 ounces in August 2000. The low was 0.16 ounces in January 1980. The ratio compares two prices, so it never says which one moved. A falling ratio can mean stocks fell, gold rose, or both. The gold-to-silver ratio carries the same limit, and no trading rule follows from either. The two-price problem is the subject of the gold-to-silver ratio.
What comes off each one
The table shows gross figures, and the deductions differ. A gold buyer pays a dealer premium going in and takes a discount coming out, plus storage and insurance (gold bar and coin value, how to sell gold). Inside an IRA, custodian and depository fees come off as well (gold IRA fees). A stock fund charges expenses, and the total return series used here does not deduct those either. Taxes diverge further. A long-term gain on gold held outside a retirement account is a collectibles gain, capped at 28% federally, while most other long-term capital gains are taxed at 0%, 15%, or 20% (IRS Topic 409). Dividends are taxed in the year they are received. The two sides’ deductions do not cancel out.
What the comparison can and cannot settle
The comparison settles what each asset returned over a stated window, how widely each moved, and how little they moved together. It cannot settle which one belongs in a portfolio. That depends on the job the money is meant to do, the time horizon, and the account that holds it. This page is research, not investment advice.
Related research
Adjacent research on this site:
- Is gold a good investment? — gold’s full inflation-adjusted record.
- Gold vs silver — the two metals on the same method.
- The gold-to-silver ratio — what a two-price ratio can and cannot say.
- Gold bar and coin value — premiums going in, discounts coming out.
- Gold IRA returns — what the metal has done inside a retirement account.
- Diversifying a retirement portfolio with a gold IRA — where the correlation question leads.
- Gold IRA vs. physical gold — the two ways to hold the metal.
- How to sell gold — the exit side of a metal position.
- Gold IRA fees — the custodian, depository, and dealer costs inside a gold IRA.
- How to sell gold — the exit side of the spread.
- Gold IRA returns — return figures for a gold IRA.
- Gold IRA vs physical gold — the account against the metal held directly.
- Diversifying retirement savings with a gold IRA — the diversification question inside a retirement account.
Frequently asked questions
Does gold perform better than the S&P 500?
It depends on the window. Measured to August 2026 in real compound annual terms, the S&P 500 leads in five of seven windows, including 11.6% against 9.0% a year from August 2016. From January 2000, gold leads, 8.1% against 5.7% a year.
Has gold outperformed the S&P 500 since 2000?
Yes, on this data. From January 2000 to August 2026, gold returned 8.1% a year after inflation, against 5.7% a year for the S&P 500 with dividends reinvested.
What is the 10-year return on gold versus the S&P 500?
Bought in August 2016 and measured to August 2026, gold returned 9.0% a year after inflation, against 11.6% a year for the S&P 500's real total return.
Is gold less volatile than the S&P 500?
Not over the full record. Annualized volatility of monthly returns from 1960 to 2026 is 15.2% for gold against 12.5% for the S&P 500's real total return. Over the last ten years the two are close, at 12.1% and 12.5%.
Does gold diversify a stock portfolio?
The two have almost no month-to-month relationship: the correlation of monthly returns is 0.00 over the full series and 0.10 over the last ten years. Correlation describes how the assets move together, not how much of each to hold, and this page gives no allocation.
Why do gold and S&P 500 comparisons disagree?
For two reasons. Some compare gold's price with the S&P 500 price index, which leaves out dividends and understates stocks. Others quote a single starting month, and the starting month alone can decide the answer.