EXACT-MONTH PRECIOUS-METAL PARITY
When did gold trade above platinum?
Both source series use USD per troy ounce. A ratio above 1 means gold's monthly average exceeded platinum's; below 1 means platinum's exceeded gold's. It is a relative-price observation, not a valuation verdict.
2026-07: $4,073.00 gold ÷ $1,622.00 platinum = 2.511.
PARITY REGIME TIMELINE
How persistent were above- and below-parity periods?
The strip preserves every month in the selected range. Gold-premium months appear above the center line; platinum-premium months appear below it. Crossings describe history, not an expected reversal.
| Decade | Months | Median ratio | Middle 50% | Gold-above share |
|---|---|---|---|---|
| 2000s | 41 | 0.552 | 0.513–0.782 | 0.0% |
| 2010s | 120 | 1.091 | 0.920–1.341 | 65.0% |
| 2020s | 79 | 2.097 | 1.847–2.388 | 100.0% |
Interpretation boundary: gold and platinum have different demand, supply and market structures. A ratio above one does not prove platinum is undervalued; a crossing does not forecast mean reversion. Monthly averages also exclude intramonth paths, dealer spreads, product premiums, storage and taxes.
READ ONE CORRECTLY
Above parity describes the quotation, not “cheap platinum”
Ratio above 1
Gold's monthly USD-per-ounce average is higher. Gold may have risen, platinum may have fallen, or both may have moved at different rates.
Ratio below 1
Platinum's monthly average is higher. Historical frequency does not establish a permanent normal level or executable arbitrage.
Ratio crosses 1
The ordering changed between monthly averages. A crossing is evidence of relative repricing, not proof of mean reversion or a forecast.
WHY THE METALS DIVERGE
Gold and platinum do not share one demand engine
Gold combines jewelry, investment, official-reserve and technology demand. Platinum combines industrial, autocatalyst, jewelry and investment uses, with different recycling and geographically concentrated mine-supply exposures. A common USD-per-ounce scale makes the price ratio easy to compute, but it does not make those economic structures interchangeable.
The regime timeline therefore answers a narrow question: how long did one monthly average remain above the other? It cannot isolate vehicle technology, mine disruptions, central-bank demand, real yields, exchange rates or substitution without additional evidence.
METHOD AND LIMITS
Exact months, one workbook, no filled gaps
GoldObserve matches gold and platinum only when both observations occur in the same calendar month in the retained World Bank Commodity Price Data workbook. Both series are nominal USD per troy ounce. Gold divided by platinum is therefore unitless; a reading of 1.25 means gold's monthly average was 25% above platinum's.
Monthly averages smooth daily and intraday paths. They are not dealer quotes, futures settlements or products a user could have transacted at without spreads, premiums, financing, storage or taxes. The latest World Bank vintage can also revise earlier observations, so the page publishes source and normalized-record hashes with every citation and CSV.
RELATED RESEARCH
Move from parity to other independently defined questions
QUESTIONS
Gold-platinum ratio FAQ
What is the gold-platinum ratio?
GoldObserve divides the World Bank monthly-average gold price by the same month's platinum price. Both are nominal USD per troy ounce, so the result is a unitless relative-price ratio.
What does a gold-platinum ratio above 1 mean?
It means gold's monthly-average USD price per troy ounce exceeded platinum's monthly average. It does not by itself prove platinum is undervalued or gold is overvalued.
Why can gold and platinum trade differently?
Gold has major monetary, reserve, jewelry and investment roles. Platinum has substantial industrial and autocatalyst exposure alongside jewelry and investment demand. Supply concentration, substitution, recycling, currencies and market structure can affect the two metals differently.
Does a parity crossing predict a reversal?
No. A crossing records a change in relative monthly averages. The duration and direction of historical regimes do not establish fair value, causation or the next price move.
Can the World Bank history change?
Yes. GoldObserve records workbook and normalized-series hashes for its retained version, while the World Bank may revise later workbook vintages. The page is latest-vintage evidence, not a point-in-time backtest.