When You Add Dividends, Stocks Beat Gold Over 20 Years

Gold has nearly tripled the S&P 500’s price-index gain over the past 20 years. The number is real, and it is also one of the most misleading statistics in long-term investing.

When investors compare precious metals to equities, they almost always reach for the wrong figures first: spot price appreciation set against a price index, ignoring the dividend engine that makes equity returns structurally different from bullion returns. That framing error quietly flatters gold and hides what an investor is actually giving up by holding it across long horizons.

This piece walks through the data on both sides using the correct benchmarks, examines what central bank gold buying really signals (and what it does not), and lays out how serious analysts think about the role precious metals play in a portfolio built for real-world volatility. The aim is a clearer picture of the trade-off, not a verdict.

The numbers that change everything when you compare gold and stocks correctly

Start with the comparison that circulates most often, because it is the one that makes gold look like a serious rival to equities. Over the latest 20-year window, tracked in US dollars to 2026, gold delivered a cumulative return of 607.7%, or roughly 10.3% annualised. Stretch the lens to 40 years and the cumulative figure reaches 871.4%, though the annualised rate drops to a more modest 5.8%.

Against the S&P 500’s price index, which rose around 504% over the same 20 years, gold looks like it won comfortably. That is the headline number retail investors keep repeating.

It is also the wrong number.

The S&P 500 price index strips out dividends, and dividends are not a rounding error. Once they are reinvested, the S&P 500’s total return over the last 20 years climbs to a cumulative 734.3%, or approximately 11.2% annualised. The apparent gap does not just close. The ranking reverses.

Silver tells a similar story with a weaker ending. Using the SLV ETF with dividends reinvested, silver compounded at 7.9% annually over the same 20 years, a cumulative total return of roughly 361.0%, well behind both gold and the broad equity market.

Then there is the extreme case of what productive capital can do when it compounds undisturbed.

Between 1965 and 2024, Berkshire Hathaway’s per-share market value delivered an overall gain of 5,502,284%, compounding at 19.9% annually, against the S&P 500’s 39,054% total return over the same span (Berkshire Hathaway 2024 Annual Report, released February 2025).

No one should expect to replicate that. Berkshire Hathaway is an outlier by any measure. But it illustrates the structural force gold can never tap: earnings reinvested into more earnings, year after year.

The correction does not make gold worthless. What it does is expose the framing: the 20-year comparison investors most often cite is an apples-to-oranges calculation that systematically overstates gold’s competitive position. Get the benchmark right, and the opportunity cost of holding bullion becomes visible for the first time. And short-term noise cuts both ways, gold was down roughly 20% year-to-date when the source data was recorded, a reminder that long-term averages hide plenty of volatility.

The same framing errors that distort gold versus equity comparisons also surface in asset class comparisons across real estate and other hard assets, where price appreciation is regularly presented without accounting for rental yield, dividend reinvestment, or total return equivalents.

Why gold cannot compound the way a productive business does

The reversal in the numbers is not the result of gold having a bad decade. It reflects a mechanical difference in what each asset is capable of producing.

A productive business, or a broad equity index made up of thousands of them, can reinvest its earnings, expand its asset base, and distribute cash to shareholders. That creates a compounding total return, where this year’s retained profit becomes next year’s larger profit base.

Gold does none of this. Its entire return is the change in the market price of a non-yielding asset. There is no earnings line, no dividend, no reinvestment, only the price a future buyer is willing to pay.

Compounding inflation operates on real purchasing power the same way compounding returns operate on nominal wealth; over long horizons, both forces are far larger than single-year figures suggest, which is why total-return benchmarks matter more than headline price appreciation when evaluating any long-duration asset.

Warren Buffett has made this argument for decades, and it is worth stating fairly. The critique is structural, not a forecast that gold’s price will fall. Gold generates no cash flow, so its value depends wholly on speculative demand rather than any internal engine of growth.

That distinction matters for how you should think about a gold position. Speculative demand rests on future buyers paying more; fundamental demand rests on earnings and cash flows the asset produces regardless of sentiment. Precious metals sit entirely on the first side of that line. They also lack financial statements, which means the fundamental valuation tools that work on equities simply do not apply.

If you hold gold as a growth asset, you are betting that sentiment stays favourable indefinitely. That is a materially different wager from owning a business that generates cash whether or not other investors happen to like it this year.

Silver’s industrial demand and why it doesn’t close the gap

Silver is often pitched as gold’s more useful cousin, because it carries genuine real-economy demand. It is used in electronics and solar panels, and that industrial pull is real.

But utility and investor return are not the same thing. Silver’s dual role, part monetary metal, part industrial input, ties its price to global manufacturing cycles and commodity futures trading. That adds volatility rather than a durable return advantage.

The punchline is in the data. Despite having more economic use than gold, silver’s 7.9% 20-year annualised return trails gold’s 10.3%. Industrial demand introduces cyclical swings; it does not build a compounding mechanism. More utility, lower return.

What central bank gold buying actually tells private investors

Central banks have been buying gold at a genuinely elevated pace, and the figures deserve to be taken seriously before they are picked apart.

In 2024, central banks and official institutions added 1,045 tonnes to global reserves, worth roughly $96 billion, the third year running that purchases topped 1,000 tonnes. That sits far above the 473-tonne annual average seen between 2010 and 2021.

Central banks added 1,045 tonnes of gold in 2024, the third consecutive year above 1,000 tonnes, more than double the 473-tonne average of the previous decade.

The pace eased in 2025 to 863 tonnes of net purchases, but that is still well above pre-2022 norms, and a record 43% of surveyed central banks said they planned to increase holdings further. China’s multi-year shift away from US Treasuries toward gold reserves has been a notable part of the trend.

China’s shift toward gold and away from US Treasuries reflects a broader set of central bank reserve functions, including settlement independence and geopolitical insulation, that have no direct parallel in private portfolio management and should not be imported wholesale into individual allocation decisions.

So the signal is real. The problem is what it means for you, and here the motives matter more than the tonnage. There are three reasons this flow is a poor guide for a private portfolio.

1. Motive mismatch. Central banks buy gold for safety, liquidity, geopolitical diversification, and to reduce US dollar exposure, not to maximise risk-adjusted returns. Their objective function is nothing like yours. As the underlying research pointedly notes, no central bankers appear among the world’s wealthiest individuals.
2. Pro-cyclical timing. Brookings and the IMF warn that central banks often buy into strength during periods of geopolitical stress. Investors who shadow these flows risk systematically buying high.
3. Opacity and idiosyncratic aims. Official purchases can be driven by national and political objectives that are neither disclosed nor relevant to a private investor’s situation.

The measured view from analysts reinforces this. Economist Barry Eichengreen has cautioned that these purchases reflect prudent systemic diversification, not an imminent monetary collapse. Goldman Sachs strategists treat central bank buying as a supportive structural demand floor while stressing that near-term prices still hinge on real yields and monetary policy, not institutional flows alone.

The elevated buying tells you something meaningful about how official institutions are managing systemic risk in a multipolar world. It tells you nothing about whether gold is cheap, fairly valued, or expensive today, which is the only question your own allocation decision actually turns on.

How serious analysts actually use gold in a portfolio

After three sections on what gold cannot do, the temptation is to write it off. That would be the wrong conclusion, and the evidence explains why.

Long-horizon equity outperformance does not settle the question of whether gold belongs in a portfolio. The relevant test is not whether gold compounds faster over 40 years, because it plainly does not. It is what gold contributes during the specific periods when equities fail.

On that test, the record is more interesting. Between 2000 and 2021, one academic study found gold returned an average of 9.2% annually against 7.39% for equities, a gap driven by gold’s tendency to hold or gain value through the dot-com bust and the global financial crisis. That is not a coincidence; it reflects gold’s low or negative correlation with stocks precisely when correlation matters most.

Those are the conditions where gold has historically earned its place:

• High-inflation periods
• Severe equity drawdowns
• Geopolitical stress events
• US dollar weakness

Read that list against the corrected long-run data and the picture calibrates. Gold is not a growth substitute. It is insurance that occasionally pays out handsomely when the rest of the portfolio is under maximum pressure.

Sizing the allocation: what the research says versus what practitioners do

How much to hold is where theory and practice diverge, and both figures are worth understanding.

Academic portfolio studies suggest gold allocations ranging from 1% to 34% can improve risk-adjusted returns, with an optimal point around 17%. Practitioner consensus, by contrast, tends to land at a more conservative 5% to 10%.

The gap is not one side being wrong. Academic models optimise on historical correlations and returns; real-world advisers discount for liquidity, transaction costs, and the behavioural reality that most investors struggle to hold a large, non-yielding position through a long equity bull run.

Academic models optimise on historical correlations, but practitioner allocations discount heavily for the behavioural reality that most investors struggle to hold a large, non-yielding position through a long equity bull run without second-guessing the thesis.

Treat this as a calibration question rather than a single correct answer. And keep the path-dependency caveat in view: any 20- or 40-year window is heavily shaped by starting valuations, inflation regimes, and crisis timing, so no single window should be treated as the final word.

Making a clearer call on gold and equities when the comparisons have been corrected

Pull the threads together and the picture is coherent, even if it stops short of a verdict.

The corrected 20-year comparison is the anchor: the S&P 500 returned 734.3% on a total-return basis against gold’s 607.7%, or 11.2% annualised versus 10.3%. Over 40 years the gap widens, with gold’s annualised rate falling to 5.8% while equities keep compounding. Add gold’s inability to reinvest earnings and the motive-mismatch problem with following central bank flows, and the case for gold as a growth engine falls apart.

The case for gold as a diversifier does not. That is the trade-off in a single sentence: holding bullion instead of equities over long horizons carries a measurable opportunity cost, while holding none leaves a portfolio fully exposed to the exact scenarios in which equities perform worst.

The conservative 5% to 10% allocation many practitioners recommend is best understood as a position that accepts both sides of that trade at once.

Before adjusting any precious metals exposure, three diagnostic questions do more work than any return chart:

• What is my allocation size relative to the research evidence, and is it deliberate or accidental?
• Am I treating gold as a growth asset or as a hedge, and does my sizing match that intent?
• Am I comparing gold to the right benchmark, total return rather than a price index?

An investor holding a calibrated position, sized against total-return data and an honest view of gold’s function, is in a far stronger place than one leaning on price-appreciation stories or institutional flow signals to justify the holding.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors.