Affiliate disclosure: White Metal Resources may earn a commission when you open an account through links on this site. This never changes what you pay or what we write. We are not financial or tax advisors. Consult a licensed advisor before making retirement decisions.
Last updated: August 5, 2026 · By White Metal Resources Editorial
Quick answer: On monthly returns from May 2006 through August 2026, gold's correlation with U.S. stocks is roughly 0.09, essentially zero, and silver's is roughly 0.27, moderately positive. Rolling 36-month correlations swing across a wide band: gold-stocks has traveled from about minus 0.27 to plus 0.44, silver-stocks from minus 0.08 to plus 0.54. In sharp equity drawdowns, gold's rolling correlation has often turned negative.
Correlation with bonds is different. Gold's monthly correlation with 7-to-10-year Treasury returns is about 0.34, and silver's is about 0.14. Neither metal is a stable substitute for a bond position. A low long-run correlation is not a promise; it is a documented pattern that moves with the cycle.
Short on time? The essentials
- Correlation measures how two return streams move together, not whether either one earns money.
- Gold's long-run correlation with U.S. stocks has been near zero, which is what makes it useful in a diversifier sleeve.
- Silver's correlation with stocks has been higher, in the 0.20 to 0.30 range, because a larger share of silver demand is industrial.
- Rolling 36-month correlations vary widely. Gold-stocks has ranged roughly from minus 0.27 to plus 0.44 across the last 17 years.
- Gold-bond correlation has been slightly positive on monthly returns and is driven more by real yields than by nominal price levels.
- Gold and silver are highly correlated with each other, near 0.80, so holding both is not the same as holding two independent hedges.
- Correlation says nothing about direction. A near-zero correlation is not a promise that metals rise when stocks fall.
- Inside a self-directed IRA, correlation still matters, but fees and dealer spreads decide whether a diversifier sleeve actually helps.
- The IRS-recognized minimums are gold .995, silver .999, and platinum or palladium .9995. That does not change what the metal does.
- The Commodity Futures Trading Commission actively polices metals futures. The 2020 JPMorgan order cited gold, silver, platinum, and palladium contracts.
Correlation is the single most misread number in the case for precious metals. Below we walk what the record actually shows for gold, silver, platinum, and palladium against U.S. stocks and U.S. Treasury bonds, using monthly return data from public FRED and ETF-proxy series. The point is to describe the pattern honestly, not to sell a hedge.
What does correlation with stocks and bonds actually mean?
Correlation measures how two return streams move together over a chosen window. A correlation of plus 1 means the two assets rise and fall in lockstep. A correlation of minus 1 means one is up when the other is down, one for one. A correlation of zero means the two return streams move without a linear link.
Correlation says nothing about direction. Two assets can both be losing money and still be uncorrelated. Two assets can both be gaining and still be uncorrelated. What correlation captures is co-movement, which is why it is central to portfolio diversification but useless as a return forecast (background: SEC investor.gov, diversification).
Time window matters. A daily correlation, a monthly correlation, and a rolling multi-year correlation can each tell a different story about the same pair of assets. The figures below use monthly log returns of adjusted closing prices, an industry-standard choice that filters out short-term noise without drifting far from the underlying series.
How does gold correlate with U.S. stocks?
On monthly log returns from May 2006 through August 2026, gold's correlation with the S&P 500 was roughly plus 0.09. That is effectively zero at the sample size, and it matches the long-run pattern published across academic and industry studies (source: FRED, GOLDAMGBD228NLBM; FRED, SP500).
The proxy here is the SPDR Gold Trust (GLD) for the metal and the SPDR S&P 500 ETF (SPY) for stocks. Both track their underlying benchmarks closely enough that a monthly correlation on adjusted closes is a fair read of the metal-versus-equity link. The academic baseline in Baur and Lucey (2010) reached similar conclusions on longer daily samples of London gold fixing and the S&P 500 (source: Baur and Lucey, 2010, "Is Gold a Hedge or a Safe Haven?").
Worth knowing: a near-zero long-run correlation is not the same as a stable near-zero correlation. Gold's link to stocks tightens and loosens with the cycle, and the value in a given quarter can be far from the long-run average.
How does silver correlate with U.S. stocks?
Silver's monthly correlation with the S&P 500 across the same window was roughly plus 0.27. That is meaningfully higher than gold's reading. The gap is consistent with silver's demand structure, where a larger share of demand is industrial and therefore tied to the same manufacturing cycle that drives equity earnings.
The Silver Institute reports that industrial uses accounted for roughly 55 to 60 percent of total silver demand in recent years, spanning solar panels, electronics, and electrical contacts (source: Silver Institute, silver supply and demand). Gold does not carry a comparable industrial-demand share, which is one reason its equity link stays looser.
A moderate positive correlation still leaves room for silver to move differently from stocks in many months, and the rolling picture in the next section shows that clearly. What it rules out is any casual claim that silver is a straightforward anti-equity hedge.
Rolling correlations move with the cycle
A single full-sample number hides most of the story. Rolling correlations, computed on a moving window of monthly returns, show how the co-movement changes across cycles. The chart below uses a 36-month (3-year) window, which is a common academic default that trades stability for responsiveness.
Gold's rolling correlation with the S&P 500 has traveled from about minus 0.27 to plus 0.44 in this sample. It sat in negative territory for stretches after the 2008 crisis and again during parts of the 2020 to 2022 window. It has been positive during periods when both stocks and gold were rising on the same monetary tailwind.
Silver's rolling correlation with the S&P 500 stayed higher than gold's for most of the window. It rarely dipped below zero and reached above plus 0.5 in some post-crisis periods, when both silver and equities responded to the same reflationary flows.
The practical read: correlation is not a fixed property of any metal. It is an average of many months, and using the long-run figure to describe next quarter's behavior is a category error.
How gold and silver correlate with U.S. Treasury bonds
Metal-to-bond correlation is where the intuition often fails. Many savers assume that a "safe" asset like gold must be closely tied to Treasury bond returns. The monthly data says otherwise. Gold's correlation with the iShares 7-10 Year Treasury Bond ETF (IEF), a common proxy for the intermediate Treasury curve, was roughly plus 0.34 across the sample. Silver's was about plus 0.14.
The academic explanation runs through real yields. Gold is a non-yielding asset, so when the real return on Treasury bonds falls (real yields decline), the opportunity cost of holding gold drops and gold tends to rise. When real yields rise, gold tends to fall (background: Federal Reserve, gold prices and inflation research note).
The observable Treasury bond price moves through both real yields and inflation expectations, so the metal-to-bond monthly return correlation is neither the same as gold's link to real yields nor a stable substitute for a bond holding. Silver, with its industrial-demand tilt, tracks the bond curve even less tightly.
| Pair | Correlation | What it says in plain terms |
|---|---|---|
| Gold vs U.S. stocks (SPY) | +0.09 | Effectively no linear link on monthly returns across the sample. |
| Silver vs U.S. stocks (SPY) | +0.27 | Moderate positive; consistent with silver's larger industrial-demand share. |
| Gold vs 7-10y Treasury (IEF) | +0.34 | Moderate positive; driven partly by shared real-yield sensitivity. |
| Silver vs 7-10y Treasury (IEF) | +0.14 | Weak positive; silver tracks industrial cycles more than the bond curve. |
| Gold vs 20y+ Treasury (TLT) | +0.25 | Similar direction to IEF, slightly lower magnitude on the long end. |
| Silver vs 20y+ Treasury (TLT) | +0.06 | Near zero on the long end of the bond curve. |
| Gold vs Silver | +0.80 | Very high; both metals share monetary and macro drivers. |
Source: monthly log returns on Yahoo Finance adjusted close for GLD, SLV, SPY, IEF, and TLT (May 2006 through August 2026). GLD proxies the London gold fixing (FRED GOLDAMGBD228NLBM); SLV proxies silver spot; SPY proxies the S&P 500 (FRED SP500). Full-sample Pearson correlations of monthly log returns, computed August 2026. Checked August 2026.
Platinum and palladium: correlations tied to industrial cycles
Platinum and palladium are platinum-group metals whose demand is dominated by industrial use, notably automotive catalytic converters, electronics, and chemical catalysis (background: USGS Mineral Commodity Summaries, platinum-group metals). Both are traded in far smaller markets than gold or silver, so their prices can swing sharply on supply news from a small number of producing regions.
Because so much of platinum and palladium demand is industrial, their correlations with broad equity indices have historically been closer to silver's than to gold's. The Commodity Futures Trading Commission oversees the U.S. futures markets for all four metals and has cited them jointly in enforcement actions (source: CFTC release 8260-20, JPMorgan spoofing order).
A precious metals IRA can hold any of the four metals at the recognized fineness minimums, but the correlation profile of a sleeve heavy in platinum or palladium is not the same as one dominated by gold. That is worth naming when a saver is weighing which metal to allocate.
How gold and silver correlate with each other
Gold and silver are very highly correlated with each other. The monthly log-return correlation in this sample was roughly plus 0.80. That is nearly as strong a co-movement as two stocks in the same sector.
The practical implication is direct: adding silver to a gold sleeve does not add a second independent diversifier. It adds ounces to a highly related exposure. Silver still brings its own industrial drivers and its own volatility profile, but it is not a second, uncorrelated hedge to gold.
Anyone weighing a mixed gold-and-silver sleeve should treat the two as one broad metals exposure with two flavors, not as a two-legged diversification story. For the fee side of that decision, see the neutral hub on gold IRA fees.
Correlation is not the same as a safe haven promise
A near-zero long-run correlation between gold and stocks is often stretched into a stronger claim: that gold "goes up when stocks go down." The record does not support that claim as a rule. Baur and Lucey (2010) drew a careful distinction between a hedge (an asset uncorrelated with another on average) and a safe haven (an asset uncorrelated or negatively correlated during a specific stress event). Gold has met the safe-haven definition in some crises and not in others.
The 2008 global financial crisis, the 2011 sovereign debt stress, and the March 2020 pandemic shock each produced different metal-and-equity co-movements. Assuming the next drawdown will look like the last one is a common sales-pitch trap. Regulators flag this pattern in dedicated bulletins (background: CFTC alert on gold and silver IRA fraud; SEC alert on self-directed IRAs).
State the case in plain terms. Gold and silver have shown useful diversification properties on average. They have not shown a guaranteed inverse move with equities in any given quarter, and no honest source claims otherwise.
What the correlation record means inside a precious metals IRA
Inside a self-directed IRA that holds physical metal, correlation still matters, but the account structure adds costs that the raw price series does not. The IRS-approved custodian, the depository, and the dealer spread all charge for the wrapper (source: IRS Investments in Collectibles Snapshot). Those costs are largely fixed dollar amounts.
A diversifier sleeve that costs 1 percent per year in fees and 3 to 5 percent in dealer spread has to earn its keep against those costs, on top of any correlation-driven benefit. On a small balance, the fees can absorb a meaningful share of the diversification story. For a step-by-step sleeve build, see diversifying a retirement portfolio with precious metals.
The IRS-recognized minimums are gold .995, silver .999, and platinum and palladium .9995 (source: 26 U.S.C. Section 408(m)). Home storage is not permitted; an approved trustee must take physical possession. Those rules do not change the correlation, but they shape the round-trip cost of expressing it inside an IRA.
How a metals sleeve responds during equity stress specifically is a related but distinct question, covered in how precious metals behave in a market downturn.
Correlation questions, answered
What is the correlation between gold and the S&P 500?
On monthly log returns from May 2006 through August 2026, gold's correlation with the S&P 500 was approximately plus 0.09, using GLD and SPY as proxies. That is effectively zero. Over longer academic samples using the London gold fixing and total-return equity indices, the figure has typically fallen between minus 0.1 and plus 0.2. The value in any given 3-year window can be well above or below that average.
Is silver negatively correlated with stocks?
No. Silver's monthly return correlation with the S&P 500 was about plus 0.27 across the sample. Silver has a much larger industrial-demand share than gold, so its price partly tracks the same manufacturing cycle that drives equity earnings. Rolling 3-year silver-stocks correlations have occasionally dipped near zero but rarely turned decisively negative.
Do gold and bonds move together?
Loosely. Gold's monthly log-return correlation with the iShares 7-10 Year Treasury Bond ETF (IEF) was about plus 0.34 across the sample. Both assets are sensitive to real yields, but the observable Treasury bond price moves through both real yields and inflation expectations. Gold is not a substitute for a bond holding; the two co-move partially, not one for one.
Is a low correlation the same as a safe haven?
No. Baur and Lucey (2010) drew a formal distinction: a hedge is uncorrelated on average, a safe haven is uncorrelated or negatively correlated during specific stress events. Gold has met the safe-haven definition in some crises and failed to meet it in others. A near-zero long-run correlation is a useful diversification input, not a promise about behavior in any given drawdown.
Which correlation window should I use to think about a metals sleeve?
For portfolio construction, rolling 3-year and rolling 5-year correlations of monthly returns are the most common academic defaults. They show how the co-movement changes across cycles and avoid over-reacting to a single quarter. A single full-sample figure is useful for a headline read but hides the dispersion that matters for planning.
Does adding silver to a gold sleeve add real diversification?
Only modestly. Gold and silver had a monthly log-return correlation of about plus 0.80 in the sample. That is very high co-movement. Silver adds different demand drivers and typically higher volatility, but it is not a second independent hedge to gold. A mixed sleeve is one broad metals exposure in two flavors, not a two-legged diversification story.
Do platinum and palladium correlate more with stocks than gold does?
They usually have. Both metals are dominated by industrial demand, so their equity correlations have historically run closer to silver's than to gold's. Their small market size also produces sharp supply-driven moves that are less common in gold. The Commodity Futures Trading Commission oversees all four metals' futures markets and has cited them jointly in enforcement actions.
Does correlation change what fees I pay inside a precious metals IRA?
No. Correlation is a property of the metal's price series. Fees are set by the custodian, the depository, and the dealer, and they run whether correlation is low or high. A near-zero correlation cannot rescue a sleeve that is too small for its fixed fee stack. For the fee framework, see the neutral hub on gold IRA fees.
Sources
- FRED, Gold Fixing Price 3:00 P.M. (London time) in London Bullion Market, U.S. Dollars per Troy Ounce (GOLDAMGBD228NLBM). Checked August 2026.
- FRED, S&P 500 Index (SP500). Checked August 2026.
- FRED, 10-Year Treasury Constant Maturity Rate (DGS10). Checked August 2026.
- Baur, D. G. and Lucey, B. M. (2010), Is Gold a Hedge or a Safe Haven? Financial Review, 45: 217-229. Checked August 2026.
- Federal Reserve Board, FEDS Notes, Gold Prices and Inflation. Checked August 2026.
- Silver Institute, Silver Supply and Demand. Checked August 2026.
- USGS Mineral Commodity Summaries, Platinum-Group Metals. Checked August 2026.
- U.S. Commodity Futures Trading Commission, Release 8260-20 (JPMorgan spoofing order, gold, silver, platinum, and palladium). Checked August 2026.
- CFTC Investor Alert, Gold and Silver IRA Fraud. Checked August 2026.
- SEC Investor Alert, Self-Directed IRAs and the Risk of Fraud. Checked August 2026.
- SEC investor.gov, Diversification. Checked August 2026.
- Cornell Legal Information Institute, 26 U.S.C. Section 408. Checked August 2026.
- IRS, Investments in Collectibles in Individually Directed Qualified Plan Accounts (Issue Snapshot). Checked August 2026.
