Gold vs the S&P 500: 50 Years of Returns, From Someone Who Isn’t Selling Gold
The short answer Over the 50 years from 1976 to 2025, $10,000 in the S&P 500 (dividends reinvested) grew to about $2.79 million. The same $10,000 in gold grew to about $309,000. That is 11.9% a year for stocks against 7.1% a year for gold, so the S&P 500 left you about nine times richer. But gold is not useless. It won 20 of those 50 years, it beat stocks for the whole decade of the 2000s, and it did better than stocks in 7 of the 9 years the S&P 500 lost money. Stocks win the marathon. Gold sometimes wins the sprint, and it can be a decent fire extinguisher. |
On January 29, 2026, gold did something it had never done before. The London morning gold price was fixed at $5,501.70 an ounce, an all-time record. It capped a year in which gold had risen about 65%, its best year since 1979.
The headlines were breathless. Gold ETFs had pulled in a record $89 billion during 2025. Friends who had never owned an ounce were asking where to buy it. The question on every burned investor's mind was simple: "Have I been in the wrong thing all along?"
Nine months later, on October 30, 2026, spot gold slipped below $3,900. That is roughly 30% below the January peak. Over the same stretch, the S&P 500 had climbed from 6,845.50 at the end of 2025 to 7,580.07, a gain of about 11% before dividends.
If you bought gold at the top in January, you have just learned something that took the generation before you 28 years to learn. By the end of this article, you will know exactly what that lesson is, and you will have the full 50-year numbers to decide for yourself.
Contents
A Quick Confession Before We Start
I don't sell gold. I don't earn a commission on coins, bars or gold IRAs, and nobody paid me to write this.
But I'm not neutral either, and you deserve to know that. I teach people to invest in low-cost index funds. So I have a bias toward the S&P 500, and I want you to watch me closely for it.
That is why this article shows you where gold won, not just where it lost. If I only showed you the flattering numbers, I would be doing exactly what I'm about to warn you about.
Who Is Telling You About Gold?
Search "gold vs S&P 500" and look at who wrote the top results. A surprising number of them are written by companies that sell gold. That doesn't make their numbers wrong. It does mean you should check which numbers they chose to show you.
Here is why that matters.
In May 2023, the US Securities and Exchange Commission charged a California gold dealer called Red Rock Secured and three of its executives. According to the SEC, the company persuaded hundreds of investors to sell the stocks and funds in their retirement accounts and buy gold and silver coins instead. Investors were told this would "protect" their retirement savings.
Investors were also told the markup on the coins would be 1% to 5%. The SEC alleged the real markups ran as high as 130%. Investors paid more than $50 million, and the SEC said Red Rock kept more than $30 million of it.
Think about what a 130% markup means. If you pay $2,300 for coins worth $1,000, gold has to more than double just for you to break even.
Ask this before you buy any gold "If I sold this back to you today, how much would you pay me?" The gap between the selling price and the buyback price is your real cost. A trustworthy dealer will answer instantly. Be wary of anyone who hesitates. |
Most gold dealers are not crooks. But a salesperson's job is to make a sale. Your job is to look at the full record, the good years and the bad. So let's do that.
The 50-Year Scoreboard: 1976 to 2025
To keep this fair, I used one independent source for both assets: the historical returns dataset kept by Professor Aswath Damodaran at NYU's Stern School of Business, last updated in January 2026. It shows the S&P 500's yearly total return (with dividends reinvested) and the yearly change in the gold price.
I took the 50 full calendar years from January 1976 to December 2025. Here is what happened to $10,000 in each.
S&P 500 (dividends reinvested) | Gold | |
$10,000 grew to | about $2,795,000 | about $309,000 |
Average yearly return | 11.9% | 7.1% |
After inflation (about 3.6% a year) | 8.0% a year | 3.4% a year |
Value in today's money | about $479,000 | about $53,000 |
Years it beat the other | 30 of 50 | 20 of 50 |
Years it lost money | 9 of 50 | 18 of 50 |
Best year | +37.2% (1995) | +126.6% (1979) |
Worst year | -36.6% (2008) | -32.6% (1981) |
Income it paid you | Dividends every year | Nothing |
The chart uses a "log scale". That simply means each step up the side is ten times bigger than the one below it, so you can see the early years clearly. On a normal scale, the stock line would look like a hockey stick and gold's line would look almost flat.
Here is the part that surprises most people. Gold lost money in 18 of those 50 years. The S&P 500 lost money in only 9. The asset people buy for "safety" went down twice as often as the one they are told is risky.
The numbers in plain words For every $1 that gold turned into over 50 years, the S&P 500 turned into about $9. After inflation, gold grew your buying power about fivefold. Stocks grew it about 48-fold. |
These figures ignore fees and taxes, which would lower both. As you'll see later, gold's costs tend to be higher, not lower.
Why Gold Can't Keep Up Over Decades: It Pays You Nothing
The biggest reason for the gap is simple. When you own the S&P 500, you own a slice of about 500 businesses. They sell things, earn profits, raise prices with inflation, and pay part of their profits to you as dividends. Reinvest those dividends, and they buy more shares, which pay more dividends.
A bar of gold does none of that. It sits in a vault. Its price only rises if someone later pays more for it than you did.
Warren Buffett made this point vividly in his 2011 letter to Berkshire Hathaway shareholders. He pointed out that all the gold in the world, about 170,000 metric tons, would fit in a cube roughly 68 feet on each side. At the time, that cube was worth about $9.6 trillion.
For the same money, he wrote, you could buy all the cropland in the United States, about 400 million acres producing roughly $200 billion of crops a year. You could also buy 16 companies the size of Exxon Mobil, and still have about $1 trillion left over.
A century later, the farmland would have produced crops every year and the companies would have paid dividends every year. The gold cube would still be exactly the same cube. In Buffett's words: "You can fondle the cube, but it will not respond."
That is the quiet engine behind the S&P 500's lead. Dividends alone made a huge difference over these 50 years. If you want to see how big, my article on dividend stocks vs index funds for retirement income walks through it.
The Oldest Trick in Investing: Pick the Start Date, Pick the Winner
Now for the most important lesson in this whole article.
Whoever chooses the starting year of a chart can make almost any asset look like the champion. Watch what happens when I keep the end date fixed at December 2025 and only move the start.
Starting in | S&P 500 per year | Gold per year | Winner |
1971 (gold standard ends) | 11.1% | 9.0% | S&P 500 |
1976 (50 years) | 11.9% | 7.1% | S&P 500 |
1980 (gold's old peak) | 12.1% | 4.8% | S&P 500, by a mile |
2000 (dot-com peak) | 8.0% | 11.0% | Gold |
2006 (20 years) | 10.9% | 11.3% | Gold, just |
2016 (10 years) | 14.7% | 15.1% | Gold, just |
Start in 1980, and gold looks like a disaster. Start in 2000, and gold looks like genius. Both charts are "true". Neither tells the whole story.
So here is a habit worth building. Whenever someone shows you a chart, a fund brochure or a social media post, look at the first date before you look at anything else. Gold sellers love charts that start in 2000 or 2001, just before stocks fell for three years in a row. Stock sellers love charts that start in 1980, the very top of gold's mania. Including people like me. That is why I've shown you both.
If you'd like to run this test on any stock or index yourself, try my free Stock Time Machine. Change the start date and watch the story change.
Decade by Decade: It's Closer Than You Think
Split the 50 years into decades and you get a real contest.
1976 to 1979: Gold, by a landslide. High inflation and an oil shock sent gold up about 38% a year.
The 1980s: Stocks, by a landslide. Gold actually lost money for the whole decade.
The 1990s: Stocks again. Gold lost money for a second decade in a row.
The 2000s: Gold. Stocks lost about 1% a year after the dot-com crash and the 2008 financial crisis, while gold made about 14% a year.
The 2010s: Stocks, comfortably. Gold made only about 3% a year.
2020 to 2025: Gold, helped by its huge 2025.
Count them up and each side wins three periods. So how does the S&P 500 end up nine times richer?
Because of what happens in the losing stretches. When stocks lost, they lost for about one decade, and businesses kept paying dividends through it. When gold lost, it lost for twenty years straight, and paid nothing while you waited.
What Happened If You Bought Gold in January 1980
Let's make that twenty-year wait real.
In 1979, gold rose about 127% in a single year. Inflation in America was running in double digits. Then, on January 21, 1980, gold hit $850 an ounce, a record.
Picture someone who finally gave in and bought that week. Everyone was talking about gold. It seemed like the safest thing in the world.
Here is what happened to that buyer:
Gold fell almost without a break through the 1980s and ended the decade a little above $400.
On July 20, 1999, it touched $252.80. That was a fall of about 70% from the peak, 19 years later.
Gold did not climb back above $850 until January 2, 2008, when it reached $859.20. That was a 28-year wait just to get the money back.
After inflation, it was far worse. According to HSBC's chief precious metals analyst, writing for the London Bullion Market Association, $850 in January 1980 is worth about $3,325 in today's money. Gold did not set a new inflation-adjusted record until April 2025. That is a 45-year wait to break even in real terms.
Notice how the same thing happened again in our own time. Gold rose about 65% in 2025, peaked in January 2026, and then fell about 30% in nine months. Nobody knows what gold does next. It could recover quickly, as it did in 2008, or slowly, as it did after 1980. But the pattern of a huge year, a January peak and a sharp drop should make anyone pause.
Fair is fair: stocks have had long waits too The S&P 500 has had painful stretches of its own. It lost about 1% a year through the whole 2000s, and Japan's stock market took 34 years to regain its 1989 peak. Read what the Nikkei's 34-year wait teaches S&P 500 investors and whether the S&P 500 has ever lost money over 20 years for the honest version of the stock story. |
Even Governments Get the Timing Wrong
Bad timing isn't just for ordinary investors. It happens to finance ministers too.
On May 7, 1999, Britain's Chancellor of the Exchequer, Gordon Brown, announced that the UK would sell about half its gold reserves. Over 17 auctions from July 1999 to March 2002, the Treasury sold about 395 tonnes at an average of roughly $275 an ounce, raising about $3.5 billion.
Gold hit its lowest point, $252.80, just two weeks after the first auction. Traders still call that low "Brown's Bottom". By 2025, the gold Britain sold would have been worth about $56 billion, according to the BBC.
To be fair, some economists have argued the sale made sense at the time. Britain swapped gold for assets that paid interest, and nobody could know in 1999 that gold was about to rise for a decade.
That is exactly the point. The man who bought at $850 in 1980 and the government that sold at $275 in 1999 made the same mistake from opposite sides. Both tried to time gold, and both got it badly wrong. If experts with a whole Treasury behind them can't time it, you and I shouldn't plan on doing it either.
The Fairest Test: Every 20- and 30-Year Period
Picking one start date is risky. So I checked every possible starting year since 1968, which is roughly when gold began trading freely, all the way through 2025.
If you held for... | S&P 500 won | Gold won |
10 years | 32 of 49 periods | 17 of 49 periods |
20 years | 27 of 39 periods | 12 of 39 periods |
30 years | 29 of 29 periods | 0 of 29 periods |
The pattern is clear. Over 10 years, gold wins about a third of the time. Over 20 years, it still wins about a third of the time, in periods that started either around 1970 or from the 1990s onward.
But over 30 years, the S&P 500 beat gold every single time. Not once, in any 30-year period since 1968, did gold come out ahead.
Why does that matter to you? Because if you're 45 today, your money probably has to last 40 years or more, through your retirement and possibly your spouse's. That is the kind of time horizon where the S&P 500's lead has been most reliable.
Where Gold Really Earns Its Keep: When Stocks Crash
Now the case for gold, and it's a real one.
In the 50 years I studied, the S&P 500 had 9 losing years. Here is how gold did in each of them.
Gold did better than stocks in 7 of those 9 years. In 2002, stocks fell 22% while gold rose 26%. In 2008, stocks fell 37% while gold rose 4%. In 2022, stocks fell 18% while gold was flat.
It isn't perfect insurance. In 1981 gold fell about 33%, far worse than stocks. In 1990 both fell about 3%.
The World Gold Council, an industry body whose members are gold mining companies, has tested what a small gold slice does to a typical portfolio. Over the 20 years to December 2025, its research found that adding 5% gold to a typical mix of stocks and bonds raised the yearly return from 6.7% to 7.0%. It also trimmed the worst fall from about 35% to about 33%.
That is useful, but notice the size of it. A small slice of gold made the ride a little smoother. It didn't transform the result. And remember who funded the study.
Here's the real value, in my view. The best portfolio is the one you won't panic out of. If a small amount of gold, or bonds, or cash, keeps you from selling everything at the bottom of a crash, it may be worth far more to you than its return. My article on why smart, hard-working people sell at the very bottom explains why that single mistake costs so much.
The Hidden Costs Nobody Puts on the Gold Chart
The 50-year numbers above assume you paid nothing to own gold. In real life, you pay, and often more than you'd pay for an index fund.
The dealer's spread. Physical coins and bars are sold above the gold price and bought back below it. On popular coins the gap may be small. On "collectible" or "special edition" coins, it can be enormous, as the Red Rock case showed.
Storage and insurance. Gold in a home safe can be stolen. Gold in a vault costs a yearly fee.
Fund fees. Gold ETFs and gold savings accounts are convenient, but they charge yearly fees too.
No income. Every year you hold gold, you give up the dividends you'd have earned from a stock index fund.
Taxes, depending on where you live. In the US, for example, physical gold and most gold ETFs are taxed as "collectibles". Long-term gains can be taxed at up to 28%, compared with a top rate of 20% on long-term stock gains. Rules differ widely by country, so check yours before you buy.
A simple rule of thumb Before buying any gold product, add up three things: what you pay over the gold price, what it costs each year to hold, and what you'd get if you sold it back tomorrow. If you can't find all three numbers in writing, don't buy it yet. |
So, Should You Own Gold? A Calm 4-Question Check
I can't tell you what to do with your money. But I can give you the questions I'd want any friend to ask before buying gold.
Question 1: Is your foundation solid?
If you don't have an emergency fund, or you're paying high interest on credit cards, fix that first. Paying off a card charging 18% is a guaranteed return gold can't match.
Question 2: Are you buying because the price just jumped?
Be honest. If gold is on the front page and everyone at dinner is talking about it, that is usually the worst time to buy. January 1980 and January 2026 both felt like the safest moment in the world to own gold.
Question 3: Is your main goal growth over 10 years or more?
If so, history suggests your core should be a low-cost, broad index fund, the kind that owns hundreds of businesses and pays you dividends while you wait. If you live outside America, here's how to invest in the S&P 500 from outside the United States.
Question 4: Would a big crash make you sell in a panic?
If yes, a smaller, steadier slice can help you hold on. That might be bonds, cash or, for some people, a little gold. Investors who use gold this way commonly keep it to 10% or less, and many set a fixed percentage and rebalance once a year. The simple lazy 3-fund portfolio shows one way to build that kind of steadiness without guesswork.
What to Do Tonight
You don't need to make any big decision today. Just do these four small things:
Find your real number. Add up everything you own in gold, including jewellery you'd genuinely sell, gold accounts and gold funds. Work out what percentage of your total savings it is.
Check the first date. Look at the last gold or stock chart someone sent you. What year did it start? Would the story change with a different start?
Ask for the buyback price. If you own physical gold, ask your dealer what they'd pay you for it today. Now you know your real cost.
Write one rule on paper. For example: "Gold will never be more than X% of my savings, and I'll check it once a year, not once a day." A written rule beats a nervous mind in a falling market.
If a crash or a past loss is what's pushing you toward gold, read what calm investors do after the market drops 20% and my plain-English guide to starting over with index funds after being burned.
Frequently Asked Questions
Has gold beaten the S&P 500 over the long run?
Not over 50 years. From 1976 to 2025, the S&P 500 with dividends reinvested returned about 11.9% a year against 7.1% for gold, turning $10,000 into about $2.79 million versus $309,000. Gold has won some shorter periods, including the 25 years from 2001 to 2025.
Has gold beaten the S&P 500 over the last 20 years?
Narrowly, yes. From 2006 to 2025 gold returned about 11.3% a year against 10.9% for the S&P 500, helped by its huge 2025. But gold then fell roughly 30% from its January 2026 peak, while the S&P 500 rose, so the 20-year race is now very close.
Is gold a good hedge against inflation?
Over very long periods gold has beaten inflation, by about 3.4% a year from 1976 to 2025. But it is unreliable over shorter spans. Someone who bought at the 1980 peak waited about 45 years to break even after inflation. The S&P 500 beat inflation by about 8% a year over the same 50 years.
How much of my portfolio should be in gold?
There is no single right answer, and it depends on your goals and nerves. Gold industry research suggests 2.5% to 10% can smooth a portfolio's ride. Many investors who own gold keep it at 10% or less and rebalance yearly. A licensed adviser can help you decide for your situation.
Why doesn't gold pay dividends?
Gold is a metal, not a business. It doesn't sell products or earn profits, so there's nothing to pay out. Its value depends only on what the next buyer will pay. Stocks represent businesses that earn profits and share them as dividends, which compound over time.
Should I sell my S&P 500 index fund to buy gold?
Be very careful about selling a long-term investment to chase an asset that just went up. History shows that buying gold after a big run, as in 1980 and early 2026, has often been followed by sharp falls. If you want some gold, many investors add a small amount gradually rather than swapping everything.
Why did gold fall so much in 2026?
By late October 2026, gold was about 30% below its January peak. Reports pointed to stubborn inflation, the chance of further US interest rate rises and selling by gold ETF investors. Higher interest rates make gold, which pays no income, less attractive to hold.
The Bottom Line
Gold is not a scam, and gold buyers are not fools. Gold has protected people in some of the worst years for stocks, and it has had brilliant decades.
But over 50 years, the S&P 500 left gold far behind. It turned $10,000 into roughly nine times as much, it won every 30-year period since 1968, and it paid you dividends the whole way.
If you've lost money before, I understand the pull of something shiny and solid you can hold in your hand. But what really protects your retirement isn't the metal. It's a simple plan, low costs, and the patience to let businesses keep working for you.
You don't have to get the timing right. You just have to stay in the game.
Free report If you've lost money in the stock market and you're wondering whether to try again, start with my free 38-page report: "Why You Lost Money in the Stock Market (And Why It Wasn't Your Fault)." It shows the hidden traps that catch ordinary investors and the calm, simple approach that works instead. |
About the Author
Stephen Loke has been a student of the market for more than 15 years. He is also an agritourism entrepreneur whose farm has been featured in Bloomberg News and Asahi Shimbun.
This article is general education, not personal financial advice. I'm not a licensed financial adviser, and I don't recommend any specific trade. Past performance doesn't guarantee future results. Please consider speaking with a licensed adviser about your own situation.
Sources
Aswath Damodaran, NYU Stern School of Business, "Historical Returns on Stocks, Bonds and Bills" (annual S&P 500 total returns and gold returns, 1928 to 2025, updated January 2026): https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/histretSP.html
US Bureau of Labor Statistics, Consumer Price Index news release, December 2025 (CPI-U 324.054): https://www.bls.gov/news.release/archives/cpi_01132026.htm
S&P Dow Jones Indices, US Equities Market Attributes, December 2025 (S&P 500 close of 6,845.50): https://www.spglobal.com/spdji/en/documents/commentary/market-attributes-us-equities-202512.pdf
Opening Bell Daily, S&P 500 in October 2026 (close of 7,580.07 on October 30, 2026): https://www.openingbelldailynews.com/p/sp-500-flat-since-may-nov-2-2026
Yahoo Finance, "Gold Price Just Hit a Yearly Low" (October 30, 2026): https://finance.yahoo.com/markets/commodities/articles/gold-price-just-hit-yearly-195453224.html
SD Bullion, daily LBMA gold price fixes for 2026 (January 29, 2026 AM fix of $5,501.70): https://sdbullion.com/gold-prices-2026
Engineering News (Reuters), "Record start to 2026 brings prospect of $5,000 gold into view" (World Gold Council ETF inflow data): https://www.engineeringnews.co.za/article/record-start-to-2026-brings-prospect-of-5-000-gold-into-view-2026-01-13
Engineering News, "Glistering start" (gold's best year since 1979): https://engineeringnews.co.za/article/glistering-start-2026-01-23
US Securities and Exchange Commission, press release 2023-93, "SEC Charges Red Rock Secured, Three Executives in Fraud Scheme Targeting Retirement Accounts" (May 15, 2023): https://www.sec.gov/news/press-release/2023-93
Berkshire Hathaway 2011 Shareholder Letter, filed with the SEC: https://www.sec.gov/Archives/edgar/data/0001067983/000119312512081960/d307508dex991.htm
James Steel (HSBC), "Echoes of the '80s? Gold Prices and Inflation Then and Now", LBMA Alchemist Issue 118: https://www.lbma.org.uk/alchemist/issue-118/echoes-of-the-80s-gold-prices-and-inflation-then-and-now
Dawn (AFP), "Gold price breaks 28-year record" (January 3, 2008): https://dawn.com/news/282812
1999 to 2002 sale of British gold reserves (summary of HM Treasury, BBC, Daily Telegraph and FT reporting): https://en.wikipedia.org/wiki/1999%E2%80%932002_sale_of_British_gold_reserves
World Gold Council, "Gold as a Strategic Asset: 2026 Edition", portfolio impact: https://www.gold.org/goldhub/research/gold-strategic-asset-2026-edition/20450
CNBC, "Gold, silver ETF owners face 28% top tax rate on profits" (March 14, 2022): https://www.cnbc.com/2022/03/14/gold-silver-etf-owners-face-28-percent-top-tax-rate-on-capital-gains.html
Comments