gold-investing · gold-vs-stocks · gold-vs-sp500
Why Gold Will Never Make You Rich (And Why You Should Still Own Some)
Gold just hit an all-time high — its first real, inflation-adjusted record since 1980. Then it fell 25% in two months.
It’s January 1980. You walk into a coin shop with $10,000 and buy gold at its all-time high: $850 an ounce. Inflation is running at 14%. The Soviets just invaded Afghanistan. Gold is the trade everyone’s chasing.
You lock it in a vault and don’t touch it.
Twenty years later you open the vault. You’ve lost 69% of your money in nominal terms — roughly 85% of your purchasing power once you adjust for inflation. Meanwhile, a co-worker who took that same $10,000 and put it into a boring S&P 500 index fund, reinvesting dividends and forgetting about it, has an account worth north of $1 million.
Same starting line. Same 20 years. Wildly different finishes.
That’s the whole thesis. But by the end, I’ll also tell you why you might still want to hold some gold anyway.
Why Gold Can’t Compound
The core idea is simpler than it sounds: compounding needs fuel, and that fuel is cash flow.
Own a share of a company and you own a sliver of a machine generating profit — some paid to you as dividends, some reinvested by the company to grow faster. Own a bond and the issuer contractually sends you a coupon on a schedule. Both of these produce cash flows. Cash flows can be reinvested. That reinvestment stacking on itself, year after year, is the entire engine of long-term wealth building.
Gold has no engine. An ounce of gold today is exactly one ounce of gold in 50 years. It doesn’t manufacture anything. It doesn’t pay a dividend. It doesn’t grow. The only path to gold making you money is someone else later paying more dollars for that same ounce — pure price appreciation, zero income, which means 100% of the return depends on getting the timing right.
Warren Buffett illustrated this with one of investing’s great thought experiments. In 2011, he pointed out that melting down every ounce of gold ever mined — all 170,000 metric tons — would form a cube about 68 feet per side, small enough to fit on a baseball infield. At that year’s price, the cube was worth $9.6 trillion. For the same money, you could instead buy every acre of US farmland plus 16 ExxonMobils, with a trillion dollars left over as pocket change.
A century from now, the farmland and the Exxons will still be generating output and dividends. The gold cube will be exactly the same size, doing nothing. As Buffett put it: you can fondle the cube, but it will not respond.
For what it’s worth, gold’s had a monster run since 2011. That same cube is worth over $24 trillion today. Gold did fine — but it did fine by repricing, not by producing anything. That distinction matters more than the headline numbers suggest.
55 Years of Data
Let’s put numbers on this. 1971 is the cleanest starting line, because that’s when Nixon ended the gold standard and let gold trade freely for the first time in decades.
From 1971 through 2024:
- S&P 500 averaged about 10.7% per year
- Gold averaged about 7.9% per year
Compounded across 50-plus years, that gap is the difference between a comfortable retirement and a dramatically more comfortable one.
Gold’s road to 7.9% was anything but smooth. It rocketed from $35 an ounce in 1971 to $850 in January 1980 — a 2,300% run fueled by double-digit inflation and oil shocks. Then it did essentially nothing for two decades. By 2000, gold had drifted down to around $265–$280, a 69% nominal loss from its peak, roughly 85% in real terms. If you bought at that high, it took until around 2024 — 44 years — for gold’s inflation-adjusted price to finally curl back to where it had started.
Stocks have never had a stretch anywhere close to that bad.
The dividend reinvestment comparison is even starker. Between 1926 and 2006 — an 80-year stretch — $10,000 in the S&P 500 grew to just over $1 million from price gains alone. But if you reinvested every dividend along the way, that same $10,000 becomes $24 million. Same market, same 80 years. The only variable is whether the cash flow got plowed back in.
That’s the whole game. And it’s a game gold can’t even enter.
The Hidden Costs That Make It Worse
The return comparison above is before tax and friction. Both of those go against gold.
The IRS treats physical gold and silver as collectibles — the same bucket as fine art. Long-term collectibles gains are taxed at a maximum rate of 28%. Stocks held over a year cap out at 20%. That’s an extra 8% of every dollar of gain handed to the government before you see a cent.
Then there’s the carrying cost. A low-cost S&P 500 index fund runs about 0.03% per year. Even GLD, the most popular gold ETF, charges 0.40% — more than 10 times that number, just to hold paper claims in a vault. If you want physical metal in hand instead, add dealer premiums on the buy, a wider spread on the sell, and if you’re paying for professional vaulting and insurance, another 0.5% to 1.5% per year for the entire duration you hold it.
None of that builds anything. It’s a toll for holding a shiny rock.
Stack the lower base return, the higher tax rate, and the ongoing storage costs together, and the real-world gap is considerably wider than the 10.7% versus 7.9% annualized returns suggest.
The Inflation Hedge That Mostly Isn’t
The common counterargument is that gold protects against inflation. The data disagrees — at least on any realistic timeline.
A widely cited Duke University study by researchers Claude Erb and Campbell Harvey, nicknamed “The Golden Dilemma,” found that over any reasonable investment horizon — 1 year, 5 years, even 20 years — gold’s price shows almost no reliable connection to actual inflation. The hedge story only holds up over horizons measured in centuries. That’s not exactly actionable for your retirement timeline.
The 2026 story makes this concrete. On January 28, 2026, gold hit an all-time high of around $5,600 an ounce — a record not just in nominal terms but finally, 46 years later, above its inflation-adjusted 1980 peak. The speed was genuinely unusual: it took gold 32 years to climb from its 1980 high to its 2011 high, then it rocketed from late 2024 to that next level in roughly 15 months.
What happened next? By mid-2026, gold had fallen about 20–25% from that all-time high, back down to the low $4,000s. The same year. Six months later. A huge chunk of the gain round-tripped in a matter of months.
JP Morgan still thinks gold is headed to $6,000 by year end — and maybe they’re right. But this is exactly the pattern Erb and Harvey’s research warned about: gold getting historically expensive, then handing a meaningful piece of it back. If you bought the euphoric top in January because it felt like the obvious safe trade, you now know exactly how the 1980 buyers felt.
The Legitimate Case for Owning Some Gold
So should you dump all your gold?
No — and here’s why. The mistake isn’t owning gold. The mistake is expecting it to grow your wealth the way stocks do.
The legitimate case for metals is insurance, not growth. Gold tends to have low — sometimes negative — correlation with stocks and bonds, especially during currency crises and geopolitical shocks. A small allocation can smooth out your worst drawdowns, even while slightly dragging your average return. It’s the same logic as car insurance: you’re not buying it because you expect the crash; you’re buying it so that if you do crash, you’re not wiped out.
Ray Dalio, who built an entire hedge fund around surviving every economic environment, generally recommends somewhere between 5% and 15% in gold, with his own all-weather portfolio sitting around 7.5%. If your plan is 5% of net worth as a disaster hedge, you’re on the conservative end of what one of the most risk-aware investors alive recommends. That’s not a fringe position.
A few other things worth knowing before you go shopping:
Physical versus paper is a real trade-off. Bars and coins carry zero counterparty risk — nobody can go bankrupt on you — but you’re responsible for storage and insurance yourself. ETFs are liquid and convenient, but you’re trusting a custodian’s vault audits. There’s no perfect answer, just a different set of trade-offs.
Silver is not gold’s quiet sibling. Over half of silver’s demand comes from industrial uses — electronics, solar panels, EVs — so it behaves more like a cyclical commodity with bigger swings in both directions. At current prices, it takes about 62 ounces of silver to equal 1 ounce of gold, and that ratio swings hard. It’s a sign silver is more speculation than gold is.
Gold isn’t guaranteed to show up exactly when you need it. During the acute panic phase of the 2008 financial crisis, gold sold off too, as investors scrambled for plain cash — before rallying hard afterward. The instant crisis insurance isn’t as automatic as the marketing suggests.
And if income is actually the goal, a boring 10-year Treasury bond pays around 4.5% right now, guaranteed every six months. If yield is what you’re after, bonds are a far more direct tool than metals.
Putting It Together
Stocks compound because they’re claims on real, growing cash flows — historically around 10 to 11% per year with dividends reinvested. Gold and silver don’t compound; they just reprice. Sometimes very well. And sometimes for 20 years straight, they go nowhere, averaging closer to 7–8% before the tax and storage drag eats even further into that.
If you want growth, that’s broad stock index funds. If you want income, that’s bonds. If you want insurance — and I think most people who watch this channel do — a modest 5 to 10% allocation to metals is reasonable and probably the right call.
Just don’t confuse the insurance policy with the retirement plan.
This is not financial advice.